E 1 P

og P
m
M 01 SI
Pr AJ
ra
2 A

The American Journal of Pathology, Vol. 178, No. 4, April 2011

C

Copyright © 2011 American Society for Investigative Pathology.

Published by Elsevier Inc. All rights reserved.
DOI: 10.1016/j.ajpath.2010.12.051

Review
Phagocytic Clearance in Neurodegeneration

Jennifer D. Sokolowski*† and James W. Mandell* Phagocytes in the CNS: Macrophages,

From the Department of Pathology * and the Medical Scientist Microglia, Astrocytes, and Neurons
Training Program and Neuroscience Graduate Program,†
University of Virginia School of Medicine, Charlottesville,
Although infiltrating macrophages and their CNS-resident
counterparts, microglia, are considered the professional
Virginia
phagocytes in the brain, there are other populations of
potential phagocytes in the CNS, including astrocytes,
neural stem cells, and possibly even neurons. These cell
The cellular and molecular mechanisms of phagocytic types derive from different lineages, exhibit different
clearance of apoptotic cells and debris have been in- characteristics, and are likely to have distinct roles in
tensely studied in invertebrate model organisms and phagocytic clearance.
in the mammalian immune system. This evolution- Microglia derive from the hematopoietic lineage, and
arily conserved process serves multiple purposes. Un- express typical pattern recognition receptors, including
cleared debris from dying cells or aggregated proteins the Toll-like receptors, Fc and complement receptors,
can be toxic and may trigger exaggerated inflamma- cytokine receptors, CD40 and MHC molecules (Table 1).
tory responses. Even though apoptotic cell death and Microglia perform typical immune cell functions, includ-
debris accumulation are key features of neurodegen- ing phagocytosis and antigen presentation, as well as
erative diseases, relatively little attention has been production of inflammatory mediators and modulation of
paid to this important homeostatic function in the the general immune response.2 Microglia are well known
central nervous system (CNS). This review attempts to for clearing dead and dying neurons after injury and
summarize our knowledge of phagocytic clearance in therefore are a prime candidate for playing a role in
the CNS, with a focus on retinal degeneration, forms phagocytic clearance in neurodegeneration.
of which are caused by mutations in genes within Astrocytes originate from neural stem cells, sharing
known phagocytic pathways, and on Alzheimer’s dis- common precursors with oligodendrocytes and neurons.
A recent microarray study on acutely isolated mouse
ease (AD). Interest in phagocytic clearance mecha-
brain astrocytes unexpectedly revealed that these cells
nisms in AD was stimulated by the discovery that
express many components of evolutionarily conserved
immunization could promote phagocytic clearance of
phagocytic pathways and numerous receptors involved
amyloid-␤; however, much less is known about clear-
in innate immunity, including Toll-like receptors, scaven-
ance of neuronal and synaptic corpses in AD and
ger receptors, and mannose receptors, as well as com-
other neurodegenerative diseases. Because the regu-
ponents of the complement system (Table 1).3 Activation
lation of phagocytic activity is intertwined with cyto- of these receptors prompts not only phagocytosis, but
kine signaling, this review also addresses the relation- also production of cytokines that lead to amplification
ships among CNS inflammation, glial responses, and and/or suppression of the immune response.4
phagocytic clearance. (Am J Pathol 2011, 178:1416 –1428; A few studies suggest that even neurons are capa-
DOI: 10.1016/j.ajpath.2010.12.051) ble of engulfment, although in many cases the phe-

Two decades of work in both Caenorhabditis elegans and Supported in part by the University of Virginia NIH-funded Medical Sci-
entist Training Program (J.D.S.) and the National Institutes of Health (NIH
Drosophila melanogaster models, as well as in mammalian
NS065447 to J.W.M.).
non-neural cells, has revealed numerous receptors and
Accepted for publication December 22, 2010.
intracellular effector molecules involved in the recognition
CME Disclosure: The authors did not disclose any relevant financial
and engulfment of apoptotic cells (Figure 1).1 However, relationships.
to what extent and in which cell types these molecules Address reprint requests to James W. Mandell, M.D., Ph.D., Depart-
function in the context of specific neurodegenerative dis- ment of Pathology, P.O. Box 800904, University of Virginia, Charlottesville,
eases is largely unstudied. VA 22908. E-mail: jwm2m@virginia.edu.

1416
 Phagocytic in Neurodegeneration 1417
AJP April 2011, Vol. 178, No. 4

Figure 1. Steps involved in phagocytosis of apoptotic debris. Apoptotic cells release diffusible find-me factors, such as ATP and chemokines, and express
cell-surface eat-me signals, including phosphatidylserine. Complement and antibody molecules serve as opsonins for debris, and bridging molecules such as milk
fat globule EGF factor 8 protein (MFGE8, also known as lactadherin) or growth arrest-specific protein 6 (GAS-6) bind to phosphatidylserine, targeting apoptotic
debris for recognition. The phagocyte can thereby identify debris either by direct recognition of phosphatidylserine through receptors such as brain-specific
angiogenesis inhibitor 1 (BAI1) or indirectly through recognition of opsonins/bridging molecules via lipoprotein receptor proteins (LRP), complement receptors,
Fc receptors, ␣v␤3/5 integrin, and the receptor tyrosine kinase Mer (MerTK), among others. Signaling via these receptors is presumed to induce phagocytosis, but
the precise players for all of these pathways have not yet been described. The small GTPase Rac1 has been implicated as a key downstream player responsible
for regulating cytoskeletal alterations that are necessary for formation of the phagocytic cup and subsequent engulfment, and at least two signaling pathways
upstream of Rac1 activation have been elucidated. First, a complex of ELMO-Dock180-CrkII acts downstream of the phosphatidylserine receptor, BAI1, and
functions as a guanine exchange factor for Rac1. Second, LRP1 interacts with GULP, an adaptor protein that has been linked to Rac1 activation, possibly via
mitogen-activated protein kinase (MAPK). Depending on the context, signaling may lead not only to phagocytosis, but also to changes in cell morphology,
induction of migration, and secretion of cytokines. For example, engulfment of apoptotic debris stimulates production of anti-inflammatory mediators, such as
transforming growth factor ␤ (TGF-␤).

nomenon is probably better characterized as endocy- study, however. When peripheral monocytes incorporate
tosis or pinocytosis. Neurons and neuronal cell lines into the parenchyma, it is usually impossible to distin-
are able to take up aggregated extracellular amyloid-␤ guish them from resident microglia morphologically, be-
(A␤) in vitro, and both low-density lipoprotein receptor- cause there are no specific markers that definitively dif-
related protein 1 (LRP1) and receptor for advanced ferentiate them. In addition, there is debate about
glycation end products (RAGE) have been implicated in whether peripheral monocytes infiltrate into the paren-
this process5–7 In addition, it has been hypothesized that chyma to a significant degree under nontraumatic cir-
neurons participate in pruning of their neighbors’ syn- cumstances.11,12 Some studies using bone marrow
apses during development. In a mouse model of prion transplant with green fluorescent protein-expressing my-
disease, ultrastructural evidence suggested that den- eloid cells suggest that, in Alzheimer’s disease models,
dritic spines enwrap degenerating presynaptic boutons.8 monocytes are recruited to the brain, evolve morpholog-
Whether neurons actually completely phagocytose and ically and functionally into microglia, and localize to sites
process the degenerating material is not clear. Although of A␤ deposition. Critics of this theory, however, have
neurons express at least some of the relevant receptors shown that the irradiation involved in bone marrow trans-
and intracellular engulfment machinery (Table 1), signal- plantation increases the leakiness of the blood-brain bar-
ing through these receptors may have different down- rier and that the previous model has not unequivocally
stream effects than in immune cells, with expression of demonstrated that infiltration occurs to a significant de-
adaptor proteins and coreceptors that are specific to gree without radiation.12
neurons.9
Although microglia are considered by some to be Activated Phagocytes: A Phenotypic Spectrum
the macrophages of the brain, it has been suggested
that peripherally derived macrophages have different Activation of a glial cell in response to debris not only
properties than resident microglia and could contribute stimulates phagocytosis, but also results in other
to clearance of A␤ plaques.10 This has been difficult to downstream effects, such as secretion of cytokines
1418 Sokolowski and Mandell
AJP April 2011, Vol. 178, No. 4

Table 1. Expression of Potential Phagocytic Receptors in Central Nervous System Cells

Expression detected
Receptor Ligand Relevance Microglia Astrocytes Neurons References

Phosphatidylserine
receptors
BAI1 PS Uptake of apoptotic cells ⫻ ⫻ ⫻ 1
TIM4 PS Uptake of apoptotic cells 1
Scavenger family PS Uptake of apoptotic cells 1, 89
receptors
Opsonin receptors
MerTK Gas6 (binds Uptake of POS, apoptotic ⫻ ⫻ 3, 26
PS) cells
Gas 6 production* ⫻ ⫻
␣v␤3/5 integrin Mfge8 (binds Uptake of apoptotic cells, ⫻ ⫻ ⫻ 3, 79
PS) uptake of A␤
MFGE8 production* ⫻
Complement receptor Complement Uptake of apoptotic cells, ⫻ ⫻ ⫻ 22, 55
uptake of A␤
Complement production* ⫻ ⫻ ⫻
FcR Immunoglobulins Uptake of apoptotic cells, ⫻ ⫻ ⫻ 64, 90
uptake of A␤
Immunoglobulin
production*
Pattern recognition
receptors
TLRs/CD14 A␤ Uptake of apoptotic cells, ⫻ ⫻ ⫻ 3, 91, 92, 93
uptake of A␤
FPRL1 A␤ Uptake of A␤ ⫻ ⫻ ⫻ 94, 95
RAGE A␤ Uptake of A␤, BBB- ⫻ ⫻ ⫻ 5, 95, 96, 97
associated A␤ receptor
CD36 scavenger A␤, ␣- Uptake of A␤ ⫻ ⫻ 42, 95, 98, 99
receptor synuclein
Lipoprotein receptors
LDLR ApoE Uptake of A␤-ApoE ⫻ ⫻ 3, 95
complexes, mediate A␤-
induced changes in ApoE
production
ApoE production* ⫻
LRP1 ApoE Uptake of A␤-ApoE ⫻ ⫻ 3, 95
complexes, BBB-
associated A␤ receptor
Uptake of A␤-ApoJ ⫻
complexes
LRP2/megalin Clusterin Clusterin production* ⫻ ⫻ ⫻ 55, 75, 95
(ApoJ)
A␤, amyloid-␤; Apo, apolipoprotein; BAI1, brain-specific angiogenesis inhibitor-1; BBB, blood-brain barrier; FcR, Fc receptor; FPRL1, formyl peptide
receptor-like 1; Gas6, growth-arrest specific 6; LDLR, low density lipoprotein receptor; LRP, low density lipoprotein receptor-related; MerTK, Mer receptor
tyrosine kinase; Mfge8, milk fat globule-EGF factor 8 (also known as lactadherin); POS, photoreceptor outer segments; PS, phosphatidylserine; RAGE,
receptor for advanced glycation endproducts; TIM4, T-cell immunoglobulin and mucin-domain-containing molecule 4; TLR, Toll-like receptor.
*Asterisk indicates that the ligand is an opsonin (bridging molecule), rather than the direct target of the phagocytic receptor.

and production of reactive oxygen species.13 An issue Classical activation has as a hallmark the production of
that complicates analysis of the literature is that re- pro-inflammatory cytokines and free radicals; alterna-
searchers refer to activation of glia without acknowl- tive activation is a less well defined anti-inflammatory
edging or accounting for the ambiguity of that label. phenotype. Expression profiling offers some utility, and
This is problematic, because there is a continuum of use of delineations that have been established for
activation for microglia that entails a range of pheno- macrophages serves as a starting point. Nonetheless,
types, and the expression of activation markers can the immunomodulatory milieu in the CNS differs, and
vary.14 For example, activation markers such as p38, activation phenotypes in the CNS may not mimic those
CD45, and FcR have been shown to be differentially in the periphery.16 Quantitative real-time PCR examin-
regulated over time in amyloid precursor protein (APP) ing expression of genes associated with inflammation
transgenic mice exposed to anti-A␤ immunotherapy.14 indicated that, in AD brain and in AD mouse models,
Using one marker is clearly insufficient to fully charac- innate immune cells exhibit a hybrid activation state
terize the activation phenotype of a glial cell. Macro- characteristic of both classical and alternative activa-
phage activation has been described in terms of clas- tion.17 Further studies are needed to generate func-
sical (M1) and alternative (M2) activation, largely tional delineations, so that researchers can better cat-
based on cytokine and receptor expression profiles.15 egorize the spectrum of glial activation phenotypes.
 Phagocytic in Neurodegeneration 1419
AJP April 2011, Vol. 178, No. 4

Regulation of Phagocytosis epithelium (RPE) is responsible for removal of this cellular

debris.25 The RPE uses the same phagocytic mecha-
Cell-cell interactions and the cytokine environment deter- nisms as are involved in the recognition and engulfment
mine whether phagocytes are ready for clearance. Sev- of apoptotic cells elsewhere in the body (Figure 1). In
eral studies have shown that astrocytes modulate micro- particular, opsonins such as milk fat globule EGF factor
glial phagocytic activity in vitro.4,18 Astrocyte-conditioned 8 protein (Mfge8; also known as lactadherin) and
medium has a general inhibitory effect,4 and a relatively growth arrest-specific protein 6 (Gas6) bind phospha-
unexplored concept is that interactions of CD40 and tidylserine and target the photoreceptor outer seg-
CD40 ligand (CD40L) could be involved.19 CD40-CD40L ments (POS) for recognition by the RPE,26,27 acting as
binding plays a role in peripheral immune activation and ligands for receptors such as the lipid scavenger re-
results in a range of outcomes, including up-regulation of ceptor CD36, an integrin adhesion receptor (␣v␤5),
costimulatory molecules, activation of antigen-presenting and MerTK (Figure 2).28,29
cells, and stimulation of cytokine production.18 Defects in clearance of POS lead to retinal degenera-
Neuronal signals also appear to influence the activa- tion characterized by photoreceptor death and progres-
tion of glia. For example, in culture, neuronal activity sive loss of vision. Due to a lack of clearance, material
modulates interferon-␥-induced major histocompatibility from the shed segments accumulates in the outer retina,
complex class II (MHCII) expression on astrocytes and and subsequently the photoreceptors and the RPE lose
microglia, ultimately dampening inflammatory activity.20 contact. The interaction between photoreceptors and the
Thus, it seems likely that neurons influence the overall RPE is critical for photoreceptor survival; on accumula-
activation phenotype and phagocytic properties of sur- tion of debris and disruption of contact, therefore, photo-
rounding glia. In the context of neurodegeneration, neu- receptor cells degenerate.25
rons may lose their ability to dampen glial proinflamma- There is evidence that POS membranes must be mod-
tory activity. ified before they are engulfed. For example, in Stargardt
T cells are able to modulate microglial activities in vitro, macular degeneration, mutation of ABCA4 results in pro-
and studies have shown that Treg cells can be neuropro- cessing defects that lead to accumulation of all-trans
tective.21 Although T cells have the machinery for direct retinal and N-retinylidene-PE on the POS membrane.
communication (eg, through CD40-CD40L interactions), Subsequent to engulfment by the RPE cells, these mole-
their relatively limited access to CNS parenchyma implies cules react and create an unmetabolizable byproduct,
that indirect mechanisms such as cytokine signaling A2E, which accrues as lipofuscin deposits. In addition,
could be involved. Further work is needed to evaluate the this byproduct can be converted to epoxides that are
complex integration of proinflammatory and anti-inflam- toxic to the RPE. The inability to degrade these products
matory cytokine signaling in glial cells and their com- and the subsequent damage renders RPE cells unable to
bined effects on phagocytic activity.22,23 support photoreceptor cells, and thus leads to retinal
degeneration.25
Clearance of Neuronal and Synaptic Corpses Retinitis pigmentosa is another retinal disorder that has
been associated with defects in phagocytic clearance.
Even though one of the key features of neurodegenera- The connection between clearance and retinitis pigmen-
tion is cell death and synapse loss, we know little about tosa was shown in an animal model, the Royal College of
the ultimate fate of cell corpses and debris in disease Surgeons (RCS) rat.30 This strain has a naturally occur-
states. Growing evidence implicates autophagy as a key ring mutation in the gene encoding MerTK, the receptor
process in several neurodegenerative diseases.24 In the- responsible for recognizing Gas6-opsonized POS and
ory, however, a neuron undergoing autophagic death inducing engulfment. Subsequent analyses identified co-
must still ultimately be cleared by phagocytosis. Although horts of patients with retinitis pigmentosa who have mu-
culture studies have convincingly shown that microglia tations in the MERTK gene.31 Of note, studies in the RCS
can ingest apoptotic neurons and neuritic blebs,13 the rat have shown that lentiviral-mediated gene therapy with
majority of the literature provides only anecdotal descrip- MerTK can correct the phagocytic defect, slow photore-
tions of the fate of degenerating neurons, either in dis- ceptor loss, and preserve retinal function for up to 7
ease or in normal aging. Direct perturbations of glial months.32 This suggests engulfment receptor defects as
phagocytic behaviors, perhaps using transgenic and a possible target for therapy of some forms of retinitis
conditional knockout mouse approaches, will be neces- pigmentosa.
sary to test cellular and molecular mechanisms of apop- To degrade debris from the shed segments, the RPE
totic neuronal clearance. must engulf and then also mediate intracellular trafficking
of the material to promote phagosome-lysosome fusion.
Retinal Degeneration Trafficking of the vesicles containing engulfed material is
critical for clearance of POS debris. Usher type 1 syn-
Phagocytic clearance plays a critical role in the mainte- drome has been associated with defects in myosin VIIa,
nance of the retinal photoreceptors, and retinal degener- and studies in knockout animals suggest that this retinal
ation can occur as a result of defects in modification, degeneration is due to disturbed vesicular trafficking and
recognition, engulfment, or degradation of photoreceptor fusion between phagosomes and lysosomes.25 The term
debris. Rod photoreceptors shed their outer segment tips phagocytic clearance refers not only to the uptake of de-
diurnally in response to light, and the retinal pigment bris, but also to the degradation of engulfed material. This
1420 Sokolowski and Mandell
AJP April 2011, Vol. 178, No. 4

1 2 3a

3b 4 5

PHOTORECEPTOR RETINAL
OUTER PIGMENT
SEGMENTS EPITHELIUM

Figure 2. Sequence of events and factors involved in clearance of photoreceptor outer segments (POS) in the retina. Photoreceptors shed their tips diurnally in
response to light, and the retinal pigment epithelium (RPE) is responsible for clearing this cellular debris.25 Retinal degeneration can occur as a result of defects
in modification, recognition, engulfment, or degradation of photoreceptor debris. 1: Membrane modification. ABCA4 functions in removal of unmetabolizable
byproducts such as all-trans retinal and N-retinylidene-PE on the POS membranes. 2: Segments are shed and expose phosphatidylserine. 3: Opsonization. 3a:
GAS-6, protein S, and MFGE8 opsonize the shed POS by binding phosphatidylserine. 3b: Tubby and tubby-like protein 1 are ligands for receptor tyrosine kinase
MerTK and promote phagocytosis via an unknown target, independent of PS. 4: Receptor recognition and engulfment. Receptors such as MerTK, lipid scavenger
receptor CD36, integrin adhesion receptor (␣v␤5), and TYRO3 recognize the shed POS and promote engulfment. 5: Intracellular trafficking of phagosomes.
Phagosome-lysosome fusion leads to degradation of engulfed material. Mutations in genes involved in these pathways that have been linked to human disease
include ABCA4 (Stargardt’s macular degeneration), MERTK (retinitis pigmentosa), and MYO7A (Usher’s syndrome).

retinal disorder highlights the fact that these two pro- sign that astrocytes and microglia are attempting to
cesses are distinct but intimately tied, and dysfunction of clear A␤, recognized as foreign. In fact, microglia and
either can result in retinal pathology. astrocytes have been shown to contain A␤ in AD brain.
These examples illuminate the fact that there are mul- This accumulation is presumed to be a result of engulf-
tiple steps involved in phagocytic clearance of cellular ment, and not due to endogenous production of A␤ by
debris, and that secondary degeneration can arise due to glia, although both microglia and astrocytes do express
genetic defects in any of these processes. It will be of the amyloid precursor protein.36 Astrocytes distant from
great interest to determine whether similar genetic de- plaques do not contain significant amounts of A␤, and
fects could be causative for any neurodegenerative or astroglial A␤ deposits are clearly spatially associated
neurodevelopmental diseases affecting the brain and with plaques.34 Evidence from electron microscopy
spinal cord. shows that microglia surround amyloid fibril deposits in
AD brain tissue, suggesting that microglia are involved in
their phagocytosis.37 Finally, postmortem immunohisto-
Alzheimer’s Disease chemistry and light microscopy have shown significant
The amyloid-␤ cascade hypothesis posits that accumu- amounts of A␤ within microglia of AD patients treated with
lation of A␤ protein is the fundamental initiator of Alzhei- immunization therapy.38
mer’s disease (AD).33 It is known that A␤ is neurotoxic in Although there is evidence that glia are capable of
itself, but in addition the glial response to A␤ can con- phagocytosing A␤, questions remain as to why clearance
tribute to its toxicity (Figure 3). The fact that pathology is inefficient in the AD brain, and researchers have at-
involves accumulation of extracellular A␤ led some re- tempted to use in vitro and ex vivo studies to understand
searchers to hypothesize that plaque formation is due to these processes. These studies have yielded contradic-
defects in phagocytic clearance of the peptide and its tory results with regard to microglial and astrocytic re-
aggregates. Studies have therefore been initiated to ex- sponses to A␤. In culture, both microglia and astrocytes
amine the role that phagocytic cells play in the develop- appear to be able to phagocytose A␤,39,40 at least when
ment of A␤ plaques and AD pathology. Although it is in the soluble form. Nonetheless, the glial response to A␤
known that reactive glia localize to plaques, possess in tissue isolated from AD brains appears to vary, and
receptors that recognize A␤, and are able to phagocy- there have been conflicting reports as to whether astro-
tose it, ultimately many questions remain about their con- cytes and microglia can clear A␤ plaques ex vivo.4,39,40
tributions to AD. The relative roles of astrocytes and microglia in A␤ clear-
ance in vivo remain to be definitively clarified.
One factor to consider when deciphering the meaning
Evidence for Phagocytosis of A␤
of studies is that the aggregation state of A␤ influences
Astrocytosis and microgliosis are invariant features of the glial response, and different forms of A␤ appear to be
severe AD,34,35 and the reactive phenotype may be a taken up by different mechanisms. Culture studies have
 Phagocytic in Neurodegeneration 1421
AJP April 2011, Vol. 178, No. 4

Figure 3. Sequence of events connecting the amyloid cascade hypothesis to putative glial cell responses. Receptors that stimulate phagocytosis are also involved
in inducing inflammatory pathways, and clearly the mechanism of activation determines the activation phenotype of the cell.14,18 It has been posited that frustrated
phagocytosis may occur when glia are unable to clear debris such as A␤.35 Persistence of extracellular debris leads to continual stimulation and production of
degrading enzymes and reactive oxygen species, as well as proinflammatory signals. This chronic inflammation creates an environment that is toxic to neurons
and promotes neurodegeneration. In theory, tipping the balance to improve phagocytic clearance and decrease secretion of proinflammatory mediators may
ameliorate A␤ pathology. The dotted lines illustrate the concept that pathology may be amplified due to a pro-inflammatory milieu and/or defects in phagocytic
clearance.

suggested that soluble A␤ is taken up by microglial pi- brains induces expression of proinflammatory genes,
nocytosis,41 whereas fibrillar A␤ is usually described as and not genes involved in phagocytosis.47 In culture, the
being phagocytosed.42 Of note, oligomeric and fibrillar addition of proinflammatory cytokines inhibits both ␤1
A␤ stimulate differential activation of microglia,43 and integrin-dependent and opsonized zymosan-stimulated
have different effects on phagocytosis.15 In particular, phagocytosis; however, anti-inflammatory cytokines are
plaque composition is variable in the AD brain,44 and mi- able to restore phagocytic activity of microglia residing in
croglia preferentially associate with dense core plaques, as a proinflammatory milieu.22 Transforming growth factor-␤
opposed to diffuse deposits.35 (TGF-␤) increased A␤ clearance in culture, and reduced
plaque burden in hAPP transgenic mice overexpressing
TGF-␤.48
Recruitment and Activation of Glia A recent study found that knockout of the cytokine
Multiple mediators play a role in recruitment and activa- receptor CX3CR1 in microglia prevents neuron loss in a
tion of glial cells in AD, including A␤ itself,18 the comple- mouse model of AD, further supporting the idea that
ment system,45 chemokines and other cytokines, and microglial response to certain cytokines has a negative
immune signaling pathways.46 These signals may arise impact in AD.38 Mechanistically, it appears that proin-
from neurons, microglia, or astrocytes in the vicinity of flammatory cytokines induce production of prostaglan-
plaques. Importantly, the characteristics of these signals dins, which act on E prostanoid receptors to inhibit
influence whether glia are phagocytic or secrete proin- phagocytosis. This is supported by the fact that antago-
flammatory cytokines. nists of prostaglandin E2 (PGE2) rescue phagocytosis,22
Amyloid-␤ triggers microglial production of proinflam- and by studies showing that knockout of the E prostanoid
matory mediators, and these mediators have been shown receptor subtype 2 increases phagocytosis of A␤.39
to inhibit phagocytosis. Real-time PCR measurements Findings from one study suggest that norepinephrine
show that A␤ treatment of microglia isolated from AD influences microglial activation toward a beneficial re-
1422 Sokolowski and Mandell
AJP April 2011, Vol. 178, No. 4

sponse. Stimulation of microglia with norepinephrine sup- plain why complement is not sufficient to stimulate clear-
pressed A␤-induced cytokine production and increased ance of plaques in the inflamed AD brain.
microglial migration and phagocytosis of A␤.49 Neurons, astrocytes, and microglia have all been re-
Interactions between astrocytes and microglia may ported to express components of the complement sys-
also be involved in regulation of phagocytosis in the AD tem,55 and complement factors have been detected in the
brain. When microglia are exposed to isolated A␤ vicinity of A␤ plaques.56,57 In vitro studies have shown that
plaques, they phagocytose and clear the cores; however, A␤ directly binds both complement C3 and C1q, and that,
coculture with astrocytes or astrocyte-conditioned me- when co-incubated in cell-free suspension, A␤ can be iso-
dium suppresses microglial phagocytosis.4 The mecha- lated along with C3 activation fragments.58 This suggests
nisms underlying this interaction are not completely clear, that A␤ is sufficient to activate the complement pathway
but it is known that exposing astrocytes to A␤ results in and may augment the inflammatory reaction or promote
up-regulation and secretion of cytokines and reactive engulfment. On complement activation, subsequent cas-
species such as tumor necrosis factor ␣ (TNF␣), IL-1␤, cades of events not only lead to glial recruitment, activa-
and nitrite,50 and these may be the key factors negatively tion, and engulfment, but also result in formation of the
regulating microglial phagocytosis. CD40-CD40 ligand membrane attack complex, which is presumed to lead to
interactions merit further investigation in the context of enhanced neurotoxicity.56 The detrimental conse-
plaque clearance. Microglia and astrocytes express quences of complement activation seem to suggest that
CD40 and CD40L, respectively, and expression of both it could be a potential target for therapeutics; however,
of these factors is increased in AD.19 In addition, there is eliminating complement C3 appears to worsen outcome,
increased soluble CD40L in plasma of AD patients, and at least in murine models of AD.
these levels may be predictive of disease progression.51 In hAPP transgenic mice, complement C3 deficiency
In vitro studies have shown that CD40 interaction with caused by overexpression of a complement inhibitor,
soluble CD40L in the presence of A␤ results in inhibition soluble complement receptor-related protein y (Scrry), or
of microglial phagocytosis and a shift toward secretion of by knockout, results in increased A␤ plaque deposi-
proinflammatory cytokines.22,52 This cytokine profile has tion.59,60 This suggests that complement may help pre-
been shown to be associated with decreased phago- vent plaque formation, perhaps acting as an opsonin for
cytic activity (Figure 3).22 The relevance of astrocytic- A␤ and targeting it for clearance by glia. Also, comple-
microglial communication via CD40L has yet to be ment C3 inactivation results in increased neuronal loss
demonstrated in vivo. and accumulation of degenerating neurons.59,60 It re-
mains to be determined whether the accumulation of
Recognition of A␤ degenerating neurons is secondary to persistence of
neurotoxic A␤ (as the authors suggested) or, alterna-
Recognition of A␤ may occur via direct receptor binding tively, could be due to loss of a direct complement-
or through opsonization by complement and immuno- mediated engulfment of degenerating neurons.
globulin. Additionally, chaperone molecules such as apo- Genome-wide association studies of late-onset spo-
lipoproteins can bind A␤ and have been implicated in radic AD found significant associations with genetic vari-
engulfment. These have respective receptors on glia, ants of a C3 receptor, CR1,54 but there are conflicting
and binding of A␤-containing complexes leads to uptake. reports regarding whether astrocytes and/or microglia
Some of these components have been associated with express CR1.55,61
AD through genome-wide association studies, including
complement receptor 1, apolipoproteins E and J, and Antibodies
phosphatidylinositol-binding clathrin assembly lymphoid
myeloid leukemia protein (PICALM).53,54 Natural antibodies to A␤ appear to exist in the blood-
stream,62 and immunization with A␤ or infusion of anti-
A␤ Receptors bodies against A␤ appears to promote clearance.63 Ex-
perimentally, antibodies have been shown to localize to
A number of receptors have been implicated in recogni- plaques in vivo and to opsonize A␤ in vitro, presumably
tion and uptake of A␤ by glia (Table 1), and many are also targeting A␤ for phagocytosis by glia.64 It is unclear why,
more generally involved in innate immune responses, as in AD, natural antibodies fail to facilitate clearance of
well as in lipoprotein recognition and uptake. It is proba- A␤ plaques. Studies on samples from individual pa-
ble that receptors that directly recognize A␤ may pro- tients have shown that the array of anti-A␤ antibodies in
mote its internalization, as well as relaying signals to different patients varies in its ability to inhibit neurotox-
modulate the inflammatory response. icity induced by A␤ in vitro. This is likely due to the fact
that they have different avidities and epitope specific-
Complement System ity,62 and it suggests that individual variation in anti-
body repertoires could play a role in susceptibility to
The complement pathway may be involved in regulating the disease.
inflammatory signaling and/or targeting material for en- The antibody response in AD patients may be en-
gulfment in AD.45 In vitro experiments have suggested hanced through immunization with A␤ or with passive
that complement-dependent phagocytic clearance is in- infusion of antibodies. A study in human patients showed
hibited by proinflammatory cytokines,22 and this may ex- that intravenous immunoglobulin infusion resulted in de-
 Phagocytic in Neurodegeneration 1423
AJP April 2011, Vol. 178, No. 4

creases in A␤ concentration in cerebrospinal fluid, with a linked to regulation of LDLR endocytosis.72 Thus, there
consequent increase in the bloodstream, perhaps sug- are multiple lines of evidence connecting ApoE and
gesting clearance from the cerebrospinal fluid and redis- LDLR to A␤ clearance in AD.
tribution of A␤ to the bloodstream; this study also showed In particular, the APOE allele encoding the apolipo-
that treatment resulted in cognitive improvement.65 Im- protein ␧4 isoform has been linked to an increased risk
portantly, immunization therapies appear to increase of sporadic AD.69 Although some studies have dem-
clearance of A␤ from the parenchyma. Postmortem im- onstrated that there are isoform-dependent differences
munohistochemistry and light microscopy on brains of in A␤ uptake, there are contradictory findings and no
AD patients immunized with A␤1– 42 showed significant overall consensus or trend. Of note, apolipoproteins
clearance of plaques in response to treatment and may also regulate the inflammatory state of the brain.
revealed that A␤ was taken up by microglia.38 This Some studies suggest that ApoE provides an overall
indicates that immunization can improve pathological anti-inflammatory effect; for example, transgenic mice
indices of AD, and it suggests that opsonization of A␤ expressing the ApoE4 isoform exhibit greater produc-
with antibodies increases phagocytosis and clearance tion of inflammatory mediators than mice expressing
of plaques. ApoE3.73
Chronic administration of anti-A␤ antibody to hAPP The lipidation state of ApoE may also modulate its
transgenic mice resulted in localization of antibodies to function. It was found that the ATP binding cassette
plaques, increased FcR expression on microglia, and led transporter A1 (ABCA1) mediates lipidation of ApoE
to a reduction in A␤ deposits due to phagocytic clear- and modulates amyloid plaque formation, and genomic
ance by microglia.64 Phagocytosis may not be the only studies have suggested that polymorphisms in ABCA1
mechanism by which antibodies promote clearance. are associated with increased risk of AD.74 This is
Studies have indicated that antibody-mediated clearance noteworthy, interesting considering the roles of ABCA
occurs even in FcR knockout animals and when antibody family members in processing of material that undergoes
lacking the Fc region is administered.66,67 Furthermore, phagocytic clearance in the retina, and specifically be-
antibody therapy has been shown in some studies to
cause mutations in the ABCA4 gene cause material to be
mediate cognitive improvement without any effect on
toxic to the engulfing cells, leading to a form of retinal
plaque clearance.68 This could be due to antibody-me-
degeneration. In theory, the effects of ABCA1 mutations
diated neutralization of the toxic effect of A␤,62 which
associated with AD could be mediated through a similar
could have a subsequent effect on inflammation and
mechanism.
cognitive function.
A role for clusterin or apolipoprotein J (ApoJ) in AD
At least in vitro, FcR-mediated phagocytic clearance
pathogenesis has been suggested,75 and it has recently
of A␤ is not affected by the presence of proinflamma-
been linked to AD in two independent genome-wide as-
tory cytokines, unlike complement-mediated clear-
sociation studies.53,54 Clusterin has been shown to be
ance.22 Perhaps immunization therapy has been so
secreted by astrocytes and increased in regions associ-
effective because it works despite a general inhibition
ated with AD pathology.75 Culture studies suggest that
of phagocytosis created by proinflammatory cytokines
in the AD brain. clusterin binds A␤ and enhances endocytosis within glial
cells via the megalin receptor, also known as LRP-2,76
which is a member of the LDLR family. In addition, clus-
Chaperone Molecules terin is a complement inhibitor (also known as comple-
Apolipoproteins are A␤-binding proteins with particular ment-associated protein SP-40,40 and as complement
relevance to phagocytic clearance. They are generally lysis inhibitor) and may suppress activity of the cytolytic
known for their role in lipid and cholesterol homeostasis, membrane attack complex.75 These findings generally
but, in addition, apolipoprotein E (ApoE)-containing lipo- suggest that clusterin plays a protective role, and future
protein particles may modulate cellular uptake of A␤ studies should examine whether the polymorphisms as-
through formation of A␤-ApoE complexes. Apoliprotein E sociated with AD have altered function that may contrib-
binds A␤, and these complexes are recognized by re- ute to pathogenesis (as has been done for the APOE
ceptors such as low-density lipoprotein receptor-related alleles).
protein (LRP) and low-density lipoprotein receptor ␣2-Macroglobulin (␣2M) is another extracellular chap-
(LDLR); ligand binding stimulates clathrin-mediated en- erone protein implicated in the binding and clearance of
docytosis. There is reason to believe that defects in these A␤; in vitro, the neurotoxicity of A␤ is abrogated by co-
mechanisms could be involved in AD, especially given incubation with ␣2M.77 Like A␤-ApoE complexes, the A␤-
that genome-wide association studies have implicated ␣2M complex acts as a ligand for LRP, and LRP medi-
variants of both ApoE and the clathrin assembly protein ates internalization. One study showed that LRP
PICALM as AD risk factors.53,69 In support of this notion, expression is required for the beneficial effect of ␣2M,
studies in transgenic mice showed that knockout of ApoE which suggests that ␣2M exerts its protective effects
reduces uptake and degradation of A␤ within astrocytes through mediating clearance of A␤ from the extracel-
in vitro,70 and overexpression of LDLR prevents amyloid lular space.77 It is possible, however, that the receptor
deposition and increases A␤ clearance.71 Of note, pre- is protective through downstream anti-inflammatory ef-
senilin, a component of the gamma secretase complex fects, and the mechanisms of neuroprotection have not
with a genetic variant linked to early-onset AD, has been been determined.
1424 Sokolowski and Mandell
AJP April 2011, Vol. 178, No. 4

Ultimately, the effects of chaperone proteins on phago- immunization and A␤ clearance; the results have been
cytic clearance may be direct, through binding A␤ and conflicting. In an APP transgenic mouse model,
promoting its uptake, or indirect, exerting effects on Tg2576, it was found that, although treatment with a
phagocytic clearance through interactions with inflamma- monoclonal antibody targeted against A␤ improved
tory pathways. cognitive measures, there was no significant reduction
in parenchymal A␤.68 In contrast, another study using
Degradation and Clearance of A␤ the same mouse model found that immunization with
anti-A␤ resulted in reduction in A␤ deposits, in addition
The mechanisms by which A␤ is normally degraded and to improved memory as measured by a Y maze task.64
cleared are still under intensive study. There is in vitro Human studies have also yielded conflicting results,
evidence that both microglia and astrocytes are capable but at least one study has shown cognitive improve-
of degrading A␤,40,41 but this ability appears to be af- ment on administration of immunoglobulin.65
fected by the inflammatory environment; for example,
interferon-␥ inhibits degradation of A␤, whereas IL-4, IL-
10, and TGF-␤ enhance degradation.23
Discussion
Even if a cell engulfs debris, phagocytosis does not
Type of Debris and Glial Activation
guarantee degradation, and in fact A␤ may be traf-
ficked to the bloodstream instead of being degraded in Engulfment of apoptotic debris induces anti-inflamma-
phagocytes. Amyloid immunotherapy is associated tory responses that result in a quiet death, free from
with a reduction in A␤ deposits in the parenchyma and inflammation and cytotoxicity to neighboring cells,1
with increases in serum A␤ levels, suggesting that similar to what is seen in clearance of POS in the retina.
transport from the parenchyma to the bloodstream oc- In contrast, materials associated with Alzheimer’s dis-
curs.64 ease, Parkinson’s disease, and prion disease, including
A␤, ␣-synuclein, neuromelanin, and prion protein appear
Neuronal Death and Synapse Loss in AD to induce a proinflammatory glial response that is neuro-
toxic.47,50,82– 84 Why this difference? Many receptors in-
The number of apoptotic cells is increased in AD brain,78 volved in recognition of these materials are pattern rec-
and although this is primarily ascribed to increased neu- ognition receptors that may have evolved to recognize
ronal death, it could also reflect deficiencies in apoptotic and respond to pathogens. A response to pathogens
clearance. Levels of the phosphatidylserine-binding op- induces phagocytes to produce cytotoxic elements
sonin Mfge8 are decreased in AD,79 and could result in aimed at killing the invader.85 Ultimately, the cell’s inflam-
persistence of apoptotic corpses that are not properly matory response to these materials appears to parallel
marked for engulfment. Synapse loss is also a well- activation that would occur in response to pathogens; this
documented feature of AD.80 It remains to be deter- reaction could be overkill and maladaptive in the case of
mined whether the molecular machinery of apoptotic neurodegenerative diseases.
cell recognition and engulfment plays a significant role Alternatively, glia may not possess the machinery to
in synapse elimination (important in both development effectively engulf and degrade these materials, resulting
and degeneration). in an overload of toxic debris. Failure to clear efficiently
may lead to frustrated phagocytosis,35 and the phago-
Does Clearance Matter? Clinical Implications cyte may ramp up the inflammatory response in the face
of continual stimulation caused by persistence of debris.
A recent report described a flow cytometric analysis to
measure A␤ phagocytosis in peripheral blood mono- Clearance: For Better or for Worse?
nuclear cells.81 The AD mononuclear cells had signif-
icantly lower phagocytic activity than in age-matched Phagocytic clearance may be considered beneficial if it
controls, making this a potentially useful clinical test. It removes toxic or inflammatory stimuli from the extracel-
will be of great interest to test also the phagocytic lular environment and prevents consequent detrimental
activity of CNS-resident cells in AD (both microglia and responses. Clearance of apoptotic cells serves as a par-
astrocytes). adigm: phagocytic removal of dying cells helps avoid an
Since the inception of the A␤ hypothesis, many have exaggerated immune response after cell death. One
suggested that stimulating clearance of A␤ could ame- might expect that a parallel model would exist in the case
liorate AD. Immunization therapies have shown some of A␤ clearance; however, we still do not have definitive
potential, and both passive transfer of immunoglobu- answers regarding i) whether phagocytic clearance is
lins and immunization with A␤ peptide or peptide frag- required to eliminate toxicity of A␤ and ii) whether engulf-
ments show promise. Initial studies in transgenic APP ment prevents or promotes detrimental effects of A␤.
mice found that immunization with A␤42 in young ani- These distinctions are important in contemplating treat-
mals prevented the development of plaques, neuritic ment strategies. In the meantime, there are many features
dystrophy, and astrogliosis and that treatment of older to consider when evaluating studies that attempt to ad-
animals reduced the extent and progression of pathol- dress these questions.
ogy.63 This led to further studies aimed at determining First, the form of A␤ (soluble, fibrillar, or complexed
whether cognitive function is improved as a result of with chaperone proteins) influences its effects, in terms
 Phagocytic in Neurodegeneration 1425
AJP April 2011, Vol. 178, No. 4

of i) toxicity, ii) capacity for phagocytic clearance, and effects, and the mechanism mediating the beneficial out-
iii) effect on inflammation.15,22,41,43 The soluble form of comes for these diseases has not been defined. NSAIDs
A␤ is more toxic to neurons than the aggregated modulate microglial cytokine expression, and could af-
forms,43 and some researchers have concluded that fect neurodegenerative diseases through their effects on
aggregation is therefore protective. Formation of the microglial phagocytosis.22
aggregated or fibrillar form may be preferable to the Finally, gene therapy has shown promise in animal
soluble form in this sense; however, it is important to models of retinal degeneration. Studies have shown
note that the fibrillar form has the propensity to cause that gene therapy can correct the phagocytic defect
glial inflammatory responses, which can then cause associated with mutations in the MerTK and Myo7a
secondary neurotoxicity. Alternatively, some studies genes, leading to slowed photoreceptor loss and pres-
have shown that A␤-chaperone protein complexes can ervation of retinal function.25,32 This suggests that
be less toxic,77 and if they do not adversely activate other engulfment pathway defects could be corrected
glia, they may be the optimal way to sequester A␤. by gene therapy. Although this may be relevant only for
Defects in chaperone systems could be involved in the conditions with monogenetic causes, it is possible that
pathogenesis of AD, especially considering that asso- its application could be more widespread, and gene
ciation studies have linked polymorphisms of chaper- therapy approaches could be used to increase phago-
one proteins and their receptors to risk of AD. Further cytic clearance in a variety of neurodegenerative dis-
studies are needed to evaluate whether formation of eases.
A␤-chaperone complexes avoids the toxicity/inflamma-
tory response observed to occur with soluble and fibril-
lar A␤ and, more important, whether phagocytosis is Conclusion
required to mediate these beneficial effects.
Studies on phagocytic clearance in the context of retinal
Second, it is unclear which makes inflammation/tox-
degeneration illuminate the fact that numerous steps in
icity worse: i) persistence of extracellular debris and
the pathway are critical for normal homeostasis. Impor-
continued stimulation of glia or ii) internalization and
tant steps include modification of debris before internal-
subsequent glial activation. Either of these could lead
ization, recognition by appropriate receptors, internaliza-
to inflammation and secondary neurotoxicity. Stimula-
tion, and trafficking through degradative pathways.
tion by extracellular debris appears sufficient to induce
Although these steps are well studied in the retina, a
glial activation and inflammation, suggesting that per-
comprehensive story has yet to emerge for clearance of
sistence of plaques is undesirable. On the other hand,
aggregates, neuronal corpses, or dysfunctional syn-
it is possible that internalization perpetuates the inflam-
apses in the context of specific neurodegenerative dis-
matory reaction to a worse degree. As an example,
eases. The use of cell type-specific and inducible knock-
exposure of astrocytes to ␣-synuclein leads to up-reg-
out mice to eliminate critical components of defined
ulation of inflammatory mediators, and the response
engulfment pathways in relevant models of neurodegen-
appears to be related to the dose of ␣-synuclein inter-
erative disease can begin to answer some of these im-
nalized.86 This supports the notion that glia sense in-
portant questions.
tracellular accumulation on phagocytosis. If this mim-
ics activity in vivo, it suggests that glial phagocytosis
may augment inflammation. Acknowledgments
We are grateful to Dr. Kodi Ravichandran (Department of
Relevance of Clearance and Therapies to Microbiology, University of Virginia) for stimulating our
Promote Phagocytosis interest in apoptotic clearance and for his generous on-
going collaboration.
Although substantial evidence indicates that defects in
clearance are causative for retinal degeneration, the im-
portance of apoptotic clearance in neurodegenerative References
diseases is much less clear. Promoting certain types of
phagocytic responses may be the key; for example, mim- 1. Elliott MR, Ravichandran KS: Clearance of apoptotic cells: implica-
tions in health and disease. J Cell Biol 2010, 189:1059 –1070
icking the anti-inflammatory effects that occur during 2. Lee CY, Landreth GE: The role of microglia in amyloid clearance from
clearance of apoptotic cells may be beneficial. Immuni- the AD brain. J Neural Transm 2010, 117:949 –960
zation studies are generally limited to models of AD, and 3. Cahoy JD, Emery B, Kaushal A, Foo LC, Zamanian JL, Christopher-
they have yielded conflicting results with regard to effi- son KS, Xing Y, Lubischer JL, Krieg PA, Krupenko SA, Thompson WJ,
Barres BA: A transcriptome database for astrocytes, neurons, and
cacy. Nonetheless, a study in human patients showed
oligodendrocytes: a new resource for understanding brain develop-
that intravenous immunoglobulin infusions promoted A␤ ment and function. J Neurosci 2008, 28:264 –278
clearance from the cerebrospinal fluid, and this was as- 4. DeWitt DA, Perry G, Cohen M, Doller C, Silver J: Astrocytes regulate
sociated with cognitive improvement.65 microglial phagocytosis of senile plaque cores of Alzheimer’s dis-
Epidemiological studies have suggested that chronic ease. Exp Neurol 1998, 149:329 –340
5. Takuma K, Fang F, Zhang W, Yan S, Fukuzaki E, Du H, Sosunov A,
use of nonsteroidal anti-inflammatory drugs (NSAIDS) McKhann G, Funatsu Y, Nakamichi N, Nagai T, Mizoguchi H, Ibi D,
may help treat or prevent progression of Alzheimer’s dis- Hori O, Ogawa S, Stern DM, Yamada K, Yan SS: RAGE-mediated
ease and Parkinson’s disease.87,88 NSAIDs have multiple signaling contributes to intraneuronal transport of amyloid-beta
1426 Sokolowski and Mandell
AJP April 2011, Vol. 178, No. 4

and neuronal dysfunction. Proc Natl Acad Sci USA 2009, 28. Ryeom SW, Silverstein RL, Scotto A, Sparrow JR: Binding of anionic
106:20021–20026 phospholipids to retinal pigment epithelium may be mediated by the
6. Hu X, Crick SL, Bu G, Frieden C, Pappu RV, Lee JM: Amyloid seeds scavenger receptor CD36. J Biol Chem 1996, 271:20536 –20539
formed by cellular uptake, concentration, and aggregation of the 29. Finnemann SC, Nandrot EF: MerTK activation during RPE phagocy-
amyloid-beta peptide. Proc Natl Acad Sci USA 2009, 106:20324 – tosis in vivo requires alphaVbeta5 integrin. Adv Exp Med Biol 2006,
20329 572:499 –503
7. Fuentealba RA, Liu Q, Zhang J, Kanekiyo T, Hu X, Lee JM, LaDu MJ, 30. Bok D, Hall MO: The role of the pigment epithelium in the etiology of
Bu G: Low-density lipoprotein receptor-related protein 1 (LRP1) me- inherited retinal dystrophy in the rat. J Cell Biol 1971, 49:664 – 682
diates neuronal Abeta42 uptake and lysosomal trafficking. PLoS One 31. Gal A, Li Y, Thompson DA, Weir J, Orth U, Jacobson SG, Apfelstedt-
2010, 5:e11884 Sylla E, Vollrath D: Mutations in MERTK, the human orthologue of the
8. Siskova Z, Page A, O’Connor V, Perry VH: Degenerating synaptic RCS rat retinal dystrophy gene, cause retinitis pigmentosa. Nat Genet
boutons in prion disease: microglia activation without synaptic strip- 2000, 26:270 –271
ping. Am J Pathol 2009, 175:1610 –1621 32. Tschernutter M, Schlichtenbrede FC, Howe S, Balaggan KS, Munro
PM, Bainbridge JW, Thrasher AJ, Smith AJ, Ali RR: Long-term pres-
9. van Noort JM, Bsibsi M: Toll-like receptors in the CNS: implications for
ervation of retinal function in the RCS rat model of retinitis pigmen-
neurodegeneration and repair. Prog Brain Res 2009, 175:139 –148
tosa following lentivirus-mediated gene therapy. Gene Ther 2005,
10. Simard AR, Soulet D, Gowing G, Julien JP, Rivest S: Bone marrow-
12:694 –701
derived microglia play a critical role in restricting senile plaque for-
33. Selkoe DJ: Toward a comprehensive theory for Alzheimer’s disease.
mation in Alzheimer’s disease. Neuron 2006, 49:489 –502
Hypothesis: Alzheimer’s disease is caused by the cerebral accumu-
11. Tsuchiya T, Park KC, Toyonaga S, Yamada SM, Nakabayashi H,
lation and cytotoxicity of amyloid beta-protein. Ann N Y Acad Sci
Nakai E, Ikawa N, Furuya M, Tominaga A, Shimizu K: Characterization
2000, 924:17–25
of microglia induced from mouse embryonic stem cells and their
34. Nagele RG, D’Andrea MR, Lee H, Venkataraman V, Wang HY: Astro-
migration into the brain parenchyma. J Neuroimmunol 2005, 160: cytes accumulate A beta 42 and give rise to astrocytic amyloid
210 –218 plaques in Alzheimer disease brains. Brain Res 2003, 971:197–209
12. Mildner A, Schmidt H, Nitsche M, Merkler D, Hanisch UK, Mack M, 35. D’Andrea MR, Cole GM, Ard MD: The microglial phagocytic role with
Heikenwalder M, Bruck W, Priller J, Prinz M: Microglia in the adult specific plaque types in the Alzheimer disease brain. Neurobiol Ag-
brain arise from Ly-6ChiCCR2⫹ monocytes only under defined host ing 2004, 25:675– 683
conditions. Nat Neurosci 2007, 10:1544 –1553 36. Haass C, Hung AY, Selkoe DJ: Processing of beta-amyloid precursor
13. Fraser DA, Pisalyaput K, Tenner AJ: C1q enhances microglial clear- protein in microglia and astrocytes favors an internal localization over
ance of apoptotic neurons and neuronal blebs, and modulates sub- constitutive secretion. J Neurosci 1991, 11:3783–3793
sequent inflammatory cytokine production. J Neurochem 2010, 112: 37. Itagaki S, McGeer PL, Akiyama H, Zhu S, Selkoe D: Relationship of
733–743 microglia and astrocytes to amyloid deposits of Alzheimer disease.
14. Morgan D, Gordon MN, Tan J, Wilcock D, Rojiani AM: Dynamic J Neuroimmunol 1989, 24:173–182
complexity of the microglial activation response in transgenic models 38. Nicoll JA, Barton E, Boche D, Neal JW, Ferrer I, Thompson P, Vla-
of amyloid deposition: implications for Alzheimer therapeutics. J Neu- chouli C, Wilkinson D, Bayer A, Games D, Seubert P, Schenk D,
ropathol Exp Neurol 2005, 64:743–753 Holmes C: Abeta species removal after abeta42 immunization. J Neu-
15. Michelucci A, Heurtaux T, Grandbarbe L, Morga E, Heuschling P: ropathol Exp Neurol 2006, 65:1040 –1048
Characterization of the microglial phenotype under specific pro-in- 39. Shie FS, Breyer RM, Montine TJ: Microglia lacking E prostanoid
flammatory and anti-inflammatory conditions: effects of oligomeric receptor subtype 2 have enhanced Abeta phagocytosis yet lack
and fibrillar amyloid-beta. J Neuroimmunol 2009, 210:3–12 Abeta-activated neurotoxicity. Am J Pathol 2005, 166:1163–1172
16. Carson MJ, Sutcliffe JG: Balancing function vs. self defense: the CNS 40. Wyss-Coray T, Loike JD, Brionne TC, Lu E, Anankov R, Yan F,
as an active regulator of immune responses. J Neurosci Res 1999, Silverstein SC, Husemann J: Adult mouse astrocytes degrade amy-
55:1– 8 loid-beta in vitro and in situ. Nat Med 2003, 9:453– 457
17. Colton CA, Mott RT, Sharpe H, Xu Q, Van Nostrand WE, Vitek MP: 41. Mandrekar S, Jiang Q, Lee CY, Koenigsknecht-Talboo J, Holtzman
Expression profiles for macrophage alternative activation genes in DM, Landreth GE: Microglia mediate the clearance of soluble Abeta
AD and in mouse models of AD. J Neuroinflammation 2006, 3:27 through fluid phase macropinocytosis. J Neurosci 2009, 29:4252–
18. Town T, Nikolic V, Tan J: The microglial “activation” continuum: from 4262
innate to adaptive responses. J Neuroinflammation 2005, 2:24 42. Koenigsknecht J, Landreth G: Microglial phagocytosis of fibrillar
19. Calingasan NY, Erdely HA, Altar CA: Identification of CD40 ligand in beta-amyloid through a beta1 integrin-dependent mechanism.
Alzheimer’s disease and in animal models of Alzheimer’s disease and J Neurosci 2004, 24:9838 –9846
brain injury. Neurobiol Aging 2002, 23:31–39 43. Sondag CM, Dhawan G, Combs CK: Beta amyloid oligomers and
fibrils stimulate differential activation of primary microglia. J Neuroin-
20. Biber K, Neumann H, Inoue K, Boddeke HW: Neuronal ‘On’ and ‘Off’
flammation 2009, 6:1
signals control microglia. Trends Neurosci 2007, 30:596 – 602
44. Pereson S, Wils H, Kleinberger G, McGowan E, Vandewoestyne M,
21. Reynolds AD, Banerjee R, Liu J, Gendelman HE, Mosley RL: Neuro-
Van Broeck B, Joris G, Cuijt I, Deforce D, Hutton M, Van Broeckhoven
protective activities of CD4⫹CD25⫹ regulatory T cells in an animal
C, Kumar-Singh S: Progranulin expression correlates with dense-core
model of Parkinson’s disease. J Leukoc Biol 2007, 82:1083–1094
amyloid plaque burden in Alzheimer disease mouse models. J Pathol
22. Koenigsknecht-Talboo J, Landreth GE: Microglial phagocytosis in-
2009, 219:173–181
duced by fibrillar beta-amyloid and IgGs are differentially regulated
45. Emmerling MR, Watson MD, Raby CA, Spiegel K: The role of com-
by proinflammatory cytokines. J Neurosci 2005, 25:8240 – 8249
plement in Alzheimer’s disease pathology. Biochim Biophys Acta
23. Yamamoto M, Kiyota T, Walsh SM, Liu J, Kipnis J, Ikezu T: Cytokine- 2000, 1502:158 –171
mediated inhibition of fibrillar amyloid-beta peptide degradation by 46. Heneka MT, O’Banion MK: Inflammatory processes in Alzheimer’s
human mononuclear phagocytes. J Immunol 2008, 181:3877–3886 disease. J Neuroimmunol 2007, 184:69 –91
24. Kohli L, Roth KA: Autophagy: cerebral home cooking. Am J Pathol 47. Walker DG, Link J, Lue LF, Dalsing-Hernandez JE, Boyes BE: Gene
2010, 176:1065–1071 expression changes by amyloid beta peptide-stimulated human post-
25. Strauss O: The retinal pigment epithelium in visual function. Physiol mortem brain microglia identify activation of multiple inflammatory
Rev 2005, 85:845– 881 processes [Erratum appeared in J Leukoc Biol 2006, 80:448]. J Leu-
26. Hall MO, Prieto AL, Obin MS, Abrams TA, Burgess BL, Heeb MJ, koc Biol 2006, 79:596 – 610
Agnew BJ: Outer segment phagocytosis by cultured retinal pigment 48. Wyss-Coray T, Lin C, Yan F, Yu GQ, Rohde M, McConlogue L,
epithelial cells requires Gas6. Exp Eye Res 2001, 73:509 –520 Masliah E, Mucke L: TGF-beta1 promotes microglial amyloid-beta
27. Nandrot EF, Anand M, Almeida D, Atabai K, Sheppard D, Finnemann clearance and reduces plaque burden in transgenic mice. Nat Med
SC: Essential role for MFG-E8 as ligand for alphavbeta5 integrin in 2001, 7:612– 618
diurnal retinal phagocytosis. Proc Natl Acad Sci USA 2007, 104: 49. Jellinger KA: Challenges in neuronal apoptosis. Curr Alzheimer Res
12005–12010 2006, 3:377–391
 Phagocytic in Neurodegeneration 1427
AJP April 2011, Vol. 178, No. 4

50. Johnstone M, Gearing AJ, Miller KM: A central role for astrocytes in beta-amyloid and increased frequency of type 4 allele in late-onset
the inflammatory response to beta-amyloid; chemokines, cytokines familial Alzheimer disease. Proc Natl Acad Sci USA 1993, 90:1977–1981
and reactive oxygen species are produced. J Neuroimmunol 1999, 70. Koistinaho M, Lin S, Wu X, Esterman M, Koger D, Hanson J, Higgs R,
93:182–193 Liu F, Malkani S, Bales KR, Paul SM: Apolipoprotein E promotes
51. Desideri G, Cipollone F, Necozione S, Marini C, Lechiara MC, Taglieri astrocyte colocalization and degradation of deposited amyloid-beta
G, Zuliani G, Fellin R, Mezzetti A, di Orio F, Ferri C: Enhanced soluble peptides. Nat Med 2004, 10:719 –726
CD40 ligand and Alzheimer’s disease: evidence of a possible patho- 71. Kim J, Castellano JM, Jiang H, Basak JM, Parsadanian M, Pham V,
genetic role. Neurobiol Aging 2008, 29:348 –356 Mason SM, Paul SM, Holtzman DM: Overexpression of low-density
52. Nikolic WV, Hou H, Town T, Zhu Y, Giunta B, Sanberg CD, Zeng J, lipoprotein receptor in the brain markedly inhibits amyloid depo-
Luo D, Ehrhart J, Mori T, Sanberg PR, Tan J: Peripherally adminis- sition and increases extracellular A beta clearance. Neuron 2009,
tered human umbilical cord blood cells reduce parenchymal and 64:632– 644
vascular beta-amyloid deposits in Alzheimer mice. Stem Cells Dev 72. Tamboli IY, Prager K, Thal DR, Thelen KM, Dewachter I, Pietrzik CU,
2008, 17:423– 439 St George-Hyslop P, Sisodia SS, De Strooper B, Heneka MT, Filippov
53. Harold D, Abraham R, Hollingworth P, Sims R, Gerrish A, Hamshere MA, Müller U, van Leuven F, Lutjohann D, Walter J: Loss of gamma-
ML, et al: Genome-wide association study identifies variants at CLU secretase function impairs endocytosis of lipoprotein particles and
and PICALM associated with Alzheimer’s disease [Erratum appeared membrane cholesterol homeostasis. J Neurosci 2008, 28:12097–
in Nat Genet 2009, 41:1156]. Nat Genet 2009, 41:1088 –1093 12106
54. Lambert JC, Heath S, Even G, Campion D, Sleegers K, Hiltunen M, 73. Colton CA, Needham LK, Brown C, Cook D, Rasheed K, Burke JR,
et al: Genome-wide association study identifies variants at CLU and Strittmatter WJ, Schmechel DE, Vitek MP: APOE genotype-specific
CR1 associated with Alzheimer’s disease. Nat Genet 2009, 41:1094 – differences in human and mouse macrophage nitric oxide produc-
1099 tion. J Neuroimmunol 2004, 147:62– 67
55. Barnum SR: Complement biosynthesis in the central nervous system. 74. Koldamova R, Fitz NF, Lefterov I: The role of ATP-binding cassette
Crit Rev Oral Biol Med 1995, 6:132–146 transporter A1 in Alzheimer’s disease and neurodegeneration.
56. Tenner AJ, Fonseca MI: The double-edged flower: roles of comple- Biochim Biophys Acta 2010, 1801:824 – 830
ment protein C1q in neurodegenerative diseases. Adv Exp Med Biol 75. Nuutinen T, Suuronen T, Kauppinen A, Salminen A: Clusterin: a
2006, 586:153–176 forgotten player in Alzheimer’s disease. Brain Res Rev 2009, 61:89 –
57. Eikelenboom P, Hack CE, Rozemuller JM, Stam FC: Complement 104
activation in amyloid plaques in Alzheimer’s dementia. Virchows Arch 76. Zlokovic BV, Martel CL, Matsubara E, McComb JG, Zheng G, Mc-
B Cell Pathol Incl Mol Pathol 1989, 56:259 –262
Cluskey RT, Frangione B, Ghiso J: Glycoprotein 330/megalin: prob-
58. Bradt BM, Kolb WP, Cooper NR: Complement-dependent proinflam-
able role in receptor-mediated transport of apolipoprotein J alone and
matory properties of the Alzheimer’s disease beta-peptide. J Exp
in a complex with Alzheimer disease amyloid beta at the blood-brain
Med 1998, 188:431– 438
and blood-cerebrospinal fluid barriers. Proc Natl Acad Sci USA 1996,
59. Wyss-Coray T, Yan F, Lin AH, Lambris JD, Alexander JJ, Quigg RJ,
93:4229 – 4234
Masliah E: Prominent neurodegeneration and increased plaque for-
77. Fabrizi C, Businaro R, Lauro GM, Fumagalli L: Role of alpha2-mac-
mation in complement-inhibited Alzheimer’s mice. Proc Natl Acad Sci
roglobulin in regulating amyloid beta-protein neurotoxicity: protective
USA 2002, 99:10837–10842
or detrimental factor? J Neurochem 2001, 78:406 – 412
60. Maier M, Peng Y, Jiang L, Seabrook TJ, Carroll MC, Lemere CA:
78. Broe M, Shepherd CE, Milward EA, Halliday GM: Relationship be-
Complement C3 deficiency leads to accelerated amyloid beta plaque
tween DNA fragmentation, morphological changes and neuronal loss
deposition and neurodegeneration and modulation of the microglia/
in Alzheimer’s disease and dementia with Lewy bodies. Acta Neuro-
macrophage phenotype in amyloid precursor protein transgenic
pathol 2001, 101:616 – 624
mice. J Neurosci 2008, 28:6333– 6341
79. Boddaert J, Kinugawa K, Lambert JC, Boukhtouche F, Zoll J, Merval
61. Singhrao SK, Neal JW, Rushmere NK, Morgan BP, Gasque P: Differ-
R, Blanc-Brude O, Mann D, Berr C, Vilar J, Garabedian B, Journiac N,
ential expression of individual complement regulators in the brain and
Charue D, Silvestre JS, Duyckaerts C, Amouyel P, Mariani J, Tedgui
choroid plexus. Lab Invest 1999, 79:1247–1259
A, Mallat Z: Evidence of a role for lactadherin in Alzheimer’s disease.
62. Szabo P, Relkin N, Weksler ME: Natural human antibodies to amyloid
beta peptide. Autoimmun Rev 2008, 7:415– 420 Am J Pathol 2007, 170:921–929
63. Schenk D, Barbour R, Dunn W, Gordon G, Grajeda H, Guido T, Hu K, 80. Scheff SW, Price DA, Schmitt FA, Mufson EJ: Hippocampal synaptic
Huang J, Johnson-Wood K, Khan K, Kholodenko D, Lee M, Liao Z, loss in early Alzheimer’s disease and mild cognitive impairment.
Lieberburg I, Motter R, Mutter L, Soriano F, Shopp G, Vasquez N, Neurobiol Aging 2006, 27:1372–1384
Vandevert C, Walker S, Wogulis M, Yednock T, Games D, Seubert P: 81. Avagyan H, Goldenson B, Tse E, Masoumi A, Porter V, Wiedau-Pazos
Immunization with amyloid-beta attenuates Alzheimer-disease-like M, Sayre J, Ong R, Mahanian M, Koo P, Bae S, Micic M, Liu PT,
pathology in the PDAPP mouse. Nature 1999, 400:173–177 Rosenthal MJ, Fiala M: Immune blood biomarkers of Alzheimer dis-
64. Wilcock DM, Rojiani A, Rosenthal A, Levkowitz G, Subbarao S, ease patients. J Neuroimmunol 2009, 210:67–72
Alamed J, Wilson D, Wilson N, Freeman MJ, Gordon MN, Morgan D: 82. McGeer PL, McGeer EG: Glial reactions in Parkinson’s disease. Mov
Passive amyloid immunotherapy clears amyloid and transiently acti- Disord 2008, 23:474 – 483
vates microglia in a transgenic mouse model of amyloid deposition. 83. Wilms H, Rosenstiel P, Sievers J, Deuschl G, Zecca L, Lucius R:
J Neurosci 2004, 24:6144 – 6151 Activation of microglia by human neuromelanin is NF-kappaB depen-
65. Relkin NR, Szabo P, Adamiak B, Burgut T, Monthe C, Lent RW, dent and involves p38 mitogen-activated protein kinase: implications
Younkin S, Younkin L, Schiff R, Weksler ME: 18-Month study of for Parkinson’s disease. FASEB J 2003, 17:500 –502
intravenous immunoglobulin for treatment of mild Alzheimer disease. 84. Brown DR, Schmidt B, Kretzschmar HA: Role of microglia and host
Neurobiol Aging 2009, 30:1728 –1736 prion protein in neurotoxicity of a prion protein fragment. Nature 1996,
66. Bacskai BJ, Kajdasz ST, McLellan ME, Games D, Seubert P, Schenk 380:345–347
D, Hyman BT: Non-Fc-mediated mechanisms are involved in clear- 85. Nagl M, Kacani L, Müllauer B, Lemberger EM, Stoiber H, Sprinzl GM,
ance of amyloid-beta in vivo by immunotherapy. J Neurosci 2002, Schennach H, Dierich MP: Phagocytosis and killing of bacteria by
22:7873–7878 professional phagocytes and dendritic cells. Clin Diagn Lab Immunol
67. Das P, Howard V, Loosbrock N, Dickson D, Murphy MP, Golde TE: 2002, 9:1165–1168
Amyloid-beta immunization effectively reduces amyloid deposition in 86. Lee HJ, Suk JE, Patrick C, Bae EJ, Cho JH, Rho S, Hwang D, Masliah
FcRgamma⫺/⫺ knock-out mice. J Neurosci 2003, 23:8532– 8538 E, Lee SJ: Direct transfer of alpha-synuclein from neuron to astroglia
68. Kotilinek LA, Bacskai B, Westerman M, Kawarabayashi T, Younkin L, causes inflammatory responses in synucleinopathies. J Biol Chem
Hyman BT, Younkin S, Ashe KH: Reversible memory loss in a mouse 2010, 285:9262–9272
transgenic model of Alzheimer’s disease. J Neurosci 2002, 22: 87. Rogers J, Kirby LC, Hempelman SR, Berry DL, McGeer PL,
6331– 6335 Kaszniak AW, Zalinski J, Cofield M, Mansukhani L, Willson P:
69. Strittmatter WJ, Saunders AM, Schmechel D, Pericak-Vance M, Enghild Clinical trial of indomethacin in Alzheimer’s disease. Neurology
J, Salvesen GS, Roses AD: Apolipoprotein E: high-avidity binding to 1993, 43:1609 –1611
1428 Sokolowski and Mandell
AJP April 2011, Vol. 178, No. 4

88. Gagne JJ, Power MC: Anti-inflammatory drugs and risk of Parkinson related protein-1 mediates amyloid-beta-mediated cell death of
disease: a meta-analysis. Neurology 2010, 74:995–1002 cerebrovascular cells. Am J Pathol 2007, 171:1989 –1999
89. Kinchen JM, Ravichandran KS: Journey to the grave: signaling events 96. Donahue JE, Flaherty SL, Johanson CE, Duncan JA 3rd, Silverberg
regulating removal of apoptotic cells. J Cell Sci 2007, 120:2143–2149 GD, Miller MC, Tavares R, Yang W, Wu Q, Sabo E, Hovanesian V,
90. Okun E, Mattson MP, Arumugam TV: Involvement of Fc receptors in Stopa EG: RAGE, LRP-1, and amyloid-beta protein in Alzheimer’s
disorders of the central nervous system. Neuromolecular Med 2009, disease. Acta Neuropathol 2006, 112:405– 415
12:164 –178 97. Deane R, Du Yan S, Submamaryan RK, LaRue B, Jovanovic S,
91. Landreth GE, Reed-Geaghan EG: Toll-like receptors in Alzheimer’s Hogg E, Welch D, Manness L, Lin C, Yu J, Zhu H, Ghiso J,
disease. Curr Top Microbiol Immunol 2009, 336:137–153 Frangione B, Stern A, Schmidt AM, Armstrong DL, Arnold B,
92. Reed-Geaghan EG, Savage JC, Hise AG, Landreth GE: CD14 and Liliensiek B, Nawroth P, Hofman F, Kindy M, Stern D, Zlokovic B:
Toll-like receptors 2 and 4 are required for fibrillar A{beta}-stimulated RAGE mediates amyloid-beta peptide transport across the blood-
microglial activation. J Neurosci 2009, 29:11982–11992 brain barrier and accumulation in brain. Nat Med 2003, 9:907–913
93. Kielian T: Overview of Toll-like receptors in the CNS. Curr Top Micro- 98. Alarcón R, Fuenzalida C, Santibáñez M, von Bernhardi R: Expression
biol Immunol 2009, 336:1–14 of scavenger receptors in glial cells. Comparing the adhesion of
94. Iribarren P, Zhou Y, Hu J, Le Y, Wang JM: Role of formyl peptide astrocytes and microglia from neonatal rats to surface-bound beta-
receptor-like 1 (FPRL1/FPR2) in mononuclear phagocyte responses amyloid. J Biol Chem 2005, 280:30406 –30415
in Alzheimer disease. Immunol Res 2005, 31:165–176 99. Paresce DM, Ghosh RN, Maxfield FR: Microglial cells internalize
95. Wilhelmus MM, Otte-Höller I, van Triel JJ, Veerhuis R, Maat-Schi- aggregates of the Alzheimer’s disease amyloid beta-protein via a
eman ML, Bu G, de Waal RM, Verbeek MM: Lipoprotein receptor- scavenger receptor. Neuron 1996, 17:553–565