R E S EARCH ART I C L E

ALZHE IMER ’S D I S EASE

http://sD

ownloaded from

Neuronal heparan sulfates promote amyloid pathologyby modulating brain amyloid-b clearance andaggregation in Alzheimer’s diseaseChia-Chen Liu,1 Na Zhao,1 Yu Yamaguchi,2 John R. Cirrito,3 Takahisa Kanekiyo,1

David M. Holtzman,3 Guojun Bu1,4*

Accumulation of amyloid-b (Ab) peptide in the brain is the first critical step in the pathogenesis of Alzheimer’sdisease (AD). Studies in humans suggest that Ab clearance from the brain is frequently impaired in late-onsetAD. Ab accumulation leads to the formation of Ab aggregates, which injure synapses and contribute to even-tual neurodegeneration. Cell surface heparan sulfates (HSs), expressed on all cell types including neurons, havebeen implicated in several features in the pathogenesis of AD including its colocalization with amyloid plaquesand modulatory role in Ab aggregation. We show that removal of neuronal HS by conditional deletion of theExt1 gene, which encodes an essential glycosyltransferase for HS biosynthesis, in postnatal neurons of amyloidmodel APP/PS1 mice led to a reduction in both Ab oligomerization and the deposition of amyloid plaques. Invivo microdialysis experiments also detected an accelerated rate of Ab clearance in the brain interstitial fluid,suggesting that neuronal HS either inhibited or represented an inefficient pathway for Ab clearance. Wefound that the amounts of various HS proteoglycans (HSPGs) were increased in postmortem human braintissues from AD patients, suggesting that this pathway may contribute directly to amyloid pathogenesis.Our findings have implications for AD pathogenesis and provide insight into therapeutic interventions targetingAb-HSPG interactions.

tm.

by guest on March 26, 2021

sciencemag.org/

INTRODUCTION

Alzheimer’s disease (AD) is the most common form of dementia inwhich amyloid plaques and neurofibrillary tangles are the major path-ological hallmarks. Mounting evidence suggests that the accumulationand aggregation of amyloid-b (Ab), the major component of amyloidplaques in the brain, is a key initiating event in the pathogenesis of AD(1, 2). Ab is generated from proteolytic processing of amyloid precursorprotein (APP) by b- and g-secretases (3, 4). Genetic and biochemicalstudies showed that early-onset forms of AD (<1%) are caused bythe inheritance of autosomal-dominant mutations that affect APPprocessing, leading to increases in Ab production or the propensityfor Ab to aggregate (5). However, much less is known about the path-ological mechanisms that modulate Ab accumulation in the late-onsetforms of AD,which account formore than 99%of cases. The amount ofAb in the brain represents the net balance of Ab production and clear-ance (6). Excess Ab rapidly aggregates to form oligomers, which havebeen shown to impair synaptic function and eventual cognitive deficits(7). Ab has a relatively short half-life in the brain, with ~1 to 2 hours inmouse interstitial fluid (ISF) and ~8 hours in human cerebrospinal fluid(CSF) (8, 9). Increasing evidence indicates that impairedAb clearance is acommonprelude to late-onsetAD (10). Thus, understanding the factorsthat regulateAb clearance is critical to illuminate the pathogenic pathwaysof AD and to design effective therapies for treating AD.

Heparan sulfate proteoglycans (HSPGs), consisting of heparan sul-fate (HS) chains covalently attached to a specific protein core, are

1Department of Neuroscience, Mayo Clinic, Jacksonville, FL 32224, USA. 2Sanford Burn-ham Prebys Medical Discovery Institute, La Jolla, CA 92037, USA. 3Department of Neurol-ogy, Hope Center for Neurological Disorders, Knight Alzheimer’s Disease Research Center,Washington University School of Medicine, St. Louis, MO 63110, USA. 4Fujian ProvincialKey Laboratory of Neurodegenerative Disease and Aging Research, Institute of Neuro-science, College of Medicine, Xiamen University, Xiamen, Fujian 361100, China.*Corresponding author. E-mail: bu.guojun@mayo.edu

www.Scien

abundant cell surface and extracellular molecules that interact witha spectrum of ligands (11, 12). A critical step in the biosynthesis ofHS is the elongation of linear polysaccharides composed of alter-nating glucuronic acid (GlcA) and N-acetylglucosamine (GlcNAc)catalyzed by the EXT1 family of molecules (13). Studies have shownthat the biosynthesis of HS is disrupted upon EXT1 deficiency, in-dicating that the glycosyltransferase activity associated with EXT1 isindispensable for HS assembly (14, 15). HSPGs are ubiquitouslyexpressed in almost all mammalian cell types and regulate a widevariety of biological processes, including embryonic development,growth factor signaling, cell proliferation, adhesion and migration,and homeostasis (12, 13). HSPGs are present in senile plaques andcerebral amyloid angiopathy (CAA) (16–19). HSPGs have beenshown to bind to Ab and accelerate its oligomerization and aggrega-tion (20–22). Also, HS mediates cellular Ab uptake, neurotoxicity,and inflammatory responses induced by Ab (23–25). These findingssuggest that HSPGs might play important roles in Ab metabolismand the pathogenesis of AD.

A recent study reported that overexpression of heparanase reducesthe amyloid burden in a mouse model, likely by modulating Ab dep-osition (26). Here, using a conditional mouse model deficient in neu-ronal HS in adult brain, we bypassed the developmental effects of HSand investigated the cell type–specific roles of HS in amyloid patho-genesis. Our findings provide in vivo evidence that neuronal HS inhib-its brain Ab clearance and promotes amyloid plaque deposition.Specifically, by using in vivo microdialysis to measure the clearancerate of soluble Ab from the brain ISF, we found that deficiency ofHS in neurons facilitated Ab clearance and reduced Ab aggregation,resulting in a reduction in amyloid plaque deposition. In addition, weshowed that several HSPG species were up-regulated in postmortemhuman brain tissues from AD patients. Together, our results shed lighton the mechanisms by which HSPGs modulate brain Ab metabolism

ceTranslationalMedicine.org 30 March 2016 Vol 8 Issue 332 332ra44 1

R E S EARCH ART I C L E

and deposition and suggest that targeting Ab-HSPG interactionsmight be a promising strategy to treat AD.

by guest on March 26, 2021

http://stm.sciencem

ag.org/D

ownloaded from

RESULTS

Neuronal HS deficiency decreases Ab and amelioratesamyloid pathology in an amyloid mouse modelThe amount of Ab in the brain is a net balance of Ab production andclearance; thus, it is important to analyze the overall impact of alteredHSPG expression on brain Ab and Ab pathology. To investigate the invivo role of HSPG in brain Abmetabolism, we conditionally disruptedthe expression of the Ext1 gene encoding the HS polymerizing en-zyme in neurons by crossing Ext1flox/flox mice (27) with a calcium/calmodulin-dependent protein kinase II (CaMKII-Cre) mice (28),and then, we further bred these mice to the APPswe/PS1DE9 (APP/PS1) amyloid mouse (29). The use of CaMKII-Cre allowed us tobypass the effects of Ext1 inactivation on embryonic brain develop-ment and to selectively study the role of HSPGs in the forebrain neu-rons of adult mice, which is one of the most vulnerable brain regions inAD. These mice developed normally without detectable morpho-logical changes in the brain, consistent with previous findings (30).We found that Ext1 expression was decreased in the cortex ofknockout mice at 12 months of age, but not in the cerebellum whereCaMKII-Cre was not active (fig. S1A). Biochemical analysis showedthat HS was significantly reduced in regions where CaMKII-Crewas active, such as cortex and hippocampus, but was unaltered inthe cerebellum (fig. S1B). The residual HS observed likely representsthe expression within glial cells or brain vasculature (17, 31, 32). AsAPP/PS1 mice develop amyloid plaques at 5 to 6 months of age,which continue to increase up to 12 months of age (33), we analyzedthe burden of amyloid deposition in these mice at 12 months of age.The brain sections of the compound mutants (APP/PS1; Ext1flox/flox;CaMKII-Cre; thereafter, APP/PS1; nExt1CKO) and their littermatecontrols (APP/PS1) were immunostained using an anti-Ab antibody(34), and the extent of Ab deposition in the cortical and hippocampalregions was quantified. The amyloid deposition was markedly de-creased in APP/PS1; nExt1CKO mice (Fig. 1A). Quantification revealedthat Ab burden in 12-month-old APP/PS1; nExt1CKO mice was aboutone-third of that in APP/PS1 littermate control mice (Fig. 1A). Previ-ous studies demonstrated that HS and HSPGs colocalize with amyloidplaques, in particular in the core of amyloid deposits and blood vessels(26, 32, 35, 36). We thus examined the association of HS or HSPGcore proteins with amyloid plaques in our mouse model. Immuno-histochemical staining showed that HS was colocalized with Congo red–positive amyloid plaques in the brain parenchyma and leptomeningealarteries in APP/PS1 mice with heparinase III treatment abolishing HSimmunoreactivity (fig. S1, C and D). As expected, the association be-tween HS and amyloid plaques was markedly decreased in APP/PS1;nExt1CKO mice in the cortex, but was not altered in the brain vesselswhere CaMKII-Cre is not active (fig. S1, C and D). Deficiency of neu-ronal HS side chains did not alter the mRNA expression of HSPGprotein backbones, including transmembrane syndecans, glycosyl-phosphatidylinositol (GPI)–linked glypicans, and extracellular matrixHSPGs (perlecan and agrin), in the brains of APP/PS1; nExt1CKOmice(fig. S2). About 20 to 40% of plaques were positive for glypican-1,syndecan-3, and agrin, whereas the colocalization between perlecan andAb was minimal (fig. S3A). We observed a reduction in the percentage

www.Scien

of amyloid plaques that were positive for various HSPG subtypes(glypican-1, syndecan-3, and agrin) in the cerebral cortex of APP/PS1;nExt1CKO mice compared to control mice (fig. S3, A and B). Agrinand perlecan, the major HSPGs in the basement membrane of the ce-rebral vasculature (35, 37), and syndecan-3 were also colocalized withAb in the leptomeningeal arteries of APP/PS1 and APP/PS1; nExt1CKO

mice (fig. S3C).To further analyze how HS affects the dynamic pools of Ab, we

fractionated cortical and hippocampal mouse brain tissues into tris-buffered saline (TBS)–soluble, detergent-soluble (TBSX), and detergent-insoluble (guanidine-HCl, GDN) fractions (38), and quantified Abby enzyme-linked immunosorbent assay (ELISA). Although the ma-jority of Ab in these mice at 12 months of age was distributed in theamyloid plaques and was fractionated in the detergent-insolublefractions, Ab40 and Ab42 in the TBSX-soluble fractions were also low-er in the brains of APP/PS1; nExt1CKO mice compared with APP/PS1controls (P < 0.01 in the cortex; P < 0.05 for Ab40 and P < 0.01 forAb42 in the hippocampus) (Fig. 1, B and C; table S2). Consistent withdecreased Ab deposition, the concentrations of insoluble Ab40 andAb42 in the guanidine fractions were very high in the APP/PS1 micein the cortex and hippocampus and were reduced in both brain regionsof APP/PS1; nExt1CKO mice (P < 0.01; Fig. 1, D and E; table S2).Oligomeric Ab has been implicated as the deleterious form of Ab pep-tide (39); thus, we examined the amount of soluble oligomeric Ab inthe TBS-soluble fraction using an ELISA specific to oligomeric Ab (40)in the cortex of these mice. We found that Ab oligomers were signif-icantly decreased in APP/PS1; nExt1CKO mice compared to APP/PS1controlmice (P < 0.01; Fig. 1F). To investigate whether deficiency of HSaffected Ab oligomerization, we further assessed the ratio of Ab oligo-mers versus total Ab in the TBS-soluble fraction of the cortex. To quan-tify total Ab, the same antibody used in oligomeric Ab ELISA was usedto captureAb followed by an antibody that recognizes themiddle regionof Ab. The ratio of Ab oligomers/total Ab was significantly reduced inAPP/PS1; nExt1CKOmice (P< 0.01; Fig. 1G), indicating a decrease of Aboligomerization in APP/PS1; nExt1CKO mouse brain. This result sug-gests that neuronal HS may promote Ab oligomerization.

HSPGs have been shown to promote Ab aggregation and fibril for-mation (16, 26, 41). Thus, we next characterized amyloid plaque loadin APP/PS1 and APP/PS1; nExt1CKO mice by staining brain sectionswith thioflavin S, a fluorescent dye that binds to amyloid fibrils. Con-sistent with the Ab immunostaining pattern, we found that deficiencyof neuronal HS led to a significant reduction in fibrillar plaques in thecortex (P < 0.01) and hippocampus (P < 0.05) of APP/PS1; nExt1CKO

mice compared to APP/PS1 controls (Fig. 1H). Together, our resultsdemonstrate that the deficiency of HS in neurons reduced Ab accu-mulation and aggregation, resulting in a decrease in amyloid plaquedeposition in APP/PS1; nExt1CKO mice.

Neuronal HS deficiency decreases neuroinflammation in anamyloid mouse modelAbnormal activation of astrocytes and microglia has been observed inthe brains of AD patients and APP transgenic mouse models (42). Toexamine the extent of astrogliosis in these mice, we immunostainedthe brain sections with anti–glial fibrillary acidic protein (GFAP) anti-body followed by quantification. The immunostaining for GFAPclearly demonstrated that activation of astrocytes was suppressed inthe brains of APP/PS1; nExt1CKO mice compared with those of APP/PS1 mice at 12 months of age (P < 0.01; Fig. 2, A to C, and fig. S4). The

ceTranslationalMedicine.org 30 March 2016 Vol 8 Issue 332 332ra44 2

R E S EARCH ART I C L E

by guest on March 26, 2021

http://stm.sciencem

ag.org/D

ownloaded from

brain sections were also immunostained with anti-ionizing calcium-binding adaptor molecule 1 (Iba1) antibody to examine the extent ofmicrogliosis. The dystrophic microglial reactivity was more evidentin APP/PS1 control mice, in which a stronger intensity and activatedmorphology of microglia were observed (P < 0.01; Fig. 2, D to F, andfig. S4). These results indicated that elimination of HS in neurons led toa reduction in plaque-associated microglial activation. Western blotanalysis also confirmed that GFAP was significantly decreased in the

www.Scien

cortex (P < 0.01) and hippocampus (P < 0.05) of APP/PS1; nExt1CKO

mice (Fig. 2G). Tumor necrosis factor–a (TNF-a), interleukin-1b (IL-1b), and IL-6 are important proinflammatory mediators in-volved in the pathogenesis of AD (43). Neuronal HS deficiency resultedin a reduction of these proinflammatory cytokines, including TNF-a,IL-1b, and IL-6 (Fig. 2H). These findings demonstrate that Ext1 in-activation in neurons decreased Ab deposition accompanied by a re-duction in neuroinflammation.

Fig. 1. Neuronal HS deficiency reduces amyloid deposition. (A) Brainsections from control (APP/PS1) and neuronal HS-deficient (APP/PS1;

of Ab40 andAb42 in TBS (B), TBSX (C), andGDN (D and E) fractionswas quan-tified by ELISA. (F) Soluble oligomeric Ab in the cortex of APP/PS1 and APP/

nExt1CKO) mice (n = 6 to 10 per group) at 12 months of age were immuno-stained with a pan-Ab antibody. Scale bar, 1 mm. The percentage of areacovered by plaques was quantified, and the plaque load was normalizedto that of APP/PS1 mice. ctrl, control. (B to E) The cortical and hippocampalbrain tissues of APP/PS1 and APP/PS1; nExt1CKOmice (n = 7 to 12 per group)at 12 months of age were fractionated into TBS-soluble, detergent-soluble(TBSX), and detergent-insoluble (guanidine-HCl, GDN) fractions. The amount

PS1; nExt1CKOmice (n=10 to 12per group). (G) Ratio of soluble oligomeric Abversus total Ab in the TBS-soluble fraction in the cortex of APP/PS1 and APP/PS1; nExt1CKOmice (n = 10 to 12 per group). (H) Quantification of thioflavinS–positive amyloid plaques in the cortex and hippocampus of APP/PS1 andAPP/PS1; nExt1CKO mice (n = 13 per group) at 12 months of age. Scale bar,100 mm. Values representmeans ± SEM. N.S., not significant; *P< 0.05; **P<0.01. Statistical analysis was performed using Student’s t test.

ceTranslationalMedicine.org 30 March 2016 Vol 8 Issue 332 332ra44 3

R E S EARCH ART I C L E

by guest on March 26, 2021

http://stm.sciencem

ag.org/D

ownloaded from

Neuronal HS deficiency increases Ab clearance in thehippocampus of amyloid model miceTo investigate the mechanism underlying the reduction of Ab pathol-ogy, we used in vivo microdialysis to dynamically assess ISF Ab me-tabolism in the hippocampus of APP/PS1; nExt1CKO mice and APP/PS1 littermates at 3 to 4 months of age. Soluble Ab in ISF has beenshown to reflect total soluble Ab in extracellular pools and is signifi-cantly correlated with the amount of Ab that eventually is deposited inthe extracellular space of the brain (8, 44, 45). Hippocampal ISF wassampled in APP/PS1; nExt1CKO mice and APP/PS1 littermates for astable baseline period, during which mice were able to freely movethroughout the experiment. To understand whether the decrease in Ab

www.Scien

in APP/PS1; nExt1CKO mice was the result of altered Ab clearance fromthe ISF, we infused a potent g-secretase inhibitor directly into the miceto rapidly block Ab production, thus allowing sensitive measurement ofthe elimination rate of Ab from the ISF (Fig. 3A and fig. S5A). ISF Ab40and Ab42 concentration gradually decreased in a time-dependent man-ner after treatment with the g-secretase inhibitor, with APP/PS1;nExt1CKO mice showing a faster decline compared with control APP/PS1 mice (Fig. 3B and fig. S5B). The half-life (t1/2) of ISF Ab40 (P <0.01) and Ab42 (P < 0.05) in the hippocampus of APP/PS1; nExt1CKO

mice was significantly shorter compared to that of APP/PS1 mice (Fig.3C and fig. S5C). These results indicate that elimination of HS in neu-rons enhanced the clearance of soluble Ab from the ISF.

Fig. 2. Deficiency of neuronal HS leads to reduced neuroinflamma-tion. (A to C) Brain sections from APP/PS1 and APP/PS1; nExt1CKO mice at

of age were immunostained with Iba1 antibody. Scale bar, 1 mm. (E) Re-presentative images of Iba1 staining in the cortex and hippocampal CA1

12 months of age were immunostained with GFAP antibody. Scale bar,1 mm. (B) Representative images of GFAP staining in the cortex and hippo-campal CA1 region. Scale bar, 100 mm. (C) Stained sections were scanned onthe Aperio slide scanner and analyzed using the ImageScope software. Thepercentage of areas covered by GFAP staining in the cortex (n = 11 to 13per group) and hippocampus (n = 7 to 10 per group) was quantified. (D toF) Brain sections from APP/PS1 and APP/PS1; nExt1CKO mice at 12 months

region. Scale bar, 100 mm. (F) The percentage of area covered by Iba1 stainingwas quantified (n = 4 per group). (G) Amount of GFAP in the cortex (n = 9 pergroup) and hippocampus (n = 8 to 9 per group) of APP/PS1 and APP/PS1;nExt1CKO mice examined by Western blot. (H) Amounts of TNF-a, IL-1b,and IL-6 in the cortex of APP/PS1 and APP/PS1; nExt1CKO mice (n = 6 to 8per group) evaluated by real-time PCR. Data representmeans ± SEM. *P< 0.05;**P < 0.01. Statistical analysis was performed using Student’s t test.

ceTranslationalMedicine.org 30 March 2016 Vol 8 Issue 332 332ra44 4

R E S EARCH ART I C L E

by guest on March 26, 2021

http://stm.sciencem

ag.org/D

ownloaded from

To examine whether Ext1 inactivation in neurons affects APP pro-cessing, total APP and APP processing products in both APP/PS1 andAPP/PS1; nExt1CKO mice were analyzed. We found that there were nosignificant differences in the amount of full-length APP, soluble formsofAPP (sAPPa and sAPPb), andC-terminal fragments (CTFs) betweenAPP/PS1 and APP/PS1; nExt1CKO mice at 12 months of age examinedby Western blot analysis (Fig. 3, D to F), indicating that deficiency ofneuronal HS in these mice did not significantly affect APP processing.Given thatHS deficiency in neurons led to decreased insoluble Ab and asubstantial reduction in amyloid plaque deposition without affectingAPP processing, we also analyzed the mRNA expression of neprilysin,insulin-degrading enzyme, matrix metalloproteinase 2 (MMP2), andMMP9, which aremajor Ab-degrading enzymes in the brain. The quan-

www.Scien

titative reverse transcription polymerase chain reaction (qRT-PCR)results showed no significant differences in the mRNA expression ofthese enzymes between APP/PS1 and APP/PS1; nExt1CKO mice (Fig.3G). Together, these results indicate that deficiency of HS in neuronsacceleratedAb clearancewithout affectingAPPprocessing or the expres-sion of major Ab-degrading enzymes in amyloid model mice.

Deficiency of neuronal HSPGs increases the formation of CAAGiven that Ab clearance was enhanced in the brains of APP/PS1;nExt1CKOmice, we thus examined whether deficiency of HS in neuronsaffected CAA formation. Ab deposition (including both Ab40 andAb42) in cortical vessels manifested as CAA was increased (P < 0.01)in the APP/PS1; nExt1CKOmice, even though the parenchymal amyloid

Fig. 3. Deficiency of neuronal HS increases ISF Ab clearance in thehippocampus of APP/PS1 mice. (A to C) APP/PS1 and APP/PS1; nExt1CKO

ual linear regressions from log (% ISF Ab40) versus time for each mouse wasused to calculate themeanhalf-life (t1/2) of elimination for Ab from the ISF. Data

mice (n = 8 per group) were analyzed at the age of 3 to 4 months. To assessAb40 half-life, the mice were treated with a g-secretase inhibitor, and the hip-pocampal ISF Ab40wasmonitored. (A) ISF Ab concentrations during the hoursof 9 to 16 after probe implantation were averaged and served as the basalamount of Ab (shown as −1 to 0 hour) before g-secretase injection. (B) Thecommon logarithm of percentage baseline ISF Ab40 concentrations versustimewas plotted. Data representmeans ± SEM. (C) The slope from the individ-

representmeans± SEM. **P<0.01. (D to F) Full-lengthAPP, sAPPa, sAPPb, CTF,and b-actin were analyzed byWestern blots in APP/PS1 and APP/PS1; nExt1CKO

mice (n= 7 to 10 per group) at 12months of age. Densitometric quantificationis expressed as mean ± SEM. (G) mRNA of insulin-degrading enzyme (IDE),neprilysin (NEP), MMP2, and MMP9 in the cortex of APP/PS1 and APP/PS1;nExt1CKOmice (n = 8 per group) evaluated by real-time PCR. Statistical analysiswas performed using Student’s t test.

ceTranslationalMedicine.org 30 March 2016 Vol 8 Issue 332 332ra44 5

R E S EARCH ART I C L E

http://stm.sc

Dow

nloaded from

plaques were substantially decreased in these mice compared to APP/PS1 controls (Fig. 4, A and B, and fig. S6). Costaining of a-smoothmuscle actin (aSMA), a marker for vascular smooth muscle cells, andAb confirmed that Ab was deposited in walls of arterioles in CAA (Fig.4C).Mounting evidence suggests that the entrapment of brainAb in theISF perivascular lymphatic drainage pathway is the main cause of CAA(46). Given that deficiency of neuronal HS increased Ab clearance andreduced amyloid deposition in the brain parenchyma, relatively moreAbmight be diverted into perivascular drainage pathways for clearancein APP/PS1; nExt1CKO mice. Because several HSPGs have been shownto be expressed by cells of the cerebral vasculature and are involved inthe pathogenesis of CAA (17, 36, 47), the increased cerebrovascular am-yloid might be attributed to the HSPGs in the cerebral vasculature.

Several classes of HSPGs are increased in humanAD brain tissuesSeveral HSPGs have been shown to colocalize with senile plaques andCAA (31, 48). To examine whether the distributions of different classesof HSPGs are altered in the pathogenesis of AD, we examined the amountsof HSPGs in the temporal cortex of human control (n = 20; averageage, 85.1 ± 5.7 years) and AD (n = 20; average age, 84.4 ± 5.2 years)brain tissues (table S1). We processed the brain tissues into TBS,detergent-soluble (TBSX), and detergent-insoluble (GDN) fractions, andthe amounts of syndecans, glypicans, perlecan, and agrin were examinedby specific ELISAs. We found that the amount of syndecan-1 was notdifferent between control and AD brain tissues (Fig. 5A). The amounts

www.Scien

of syndecan-3 and syndecan-4 in both the detergent-soluble (P < 0.01for syndecan-3; P < 0.05 for syndecan-4) and detergent-insolublefractions (P < 0.05 for syndecan-3; P < 0.05 for syndecan-4) were sig-nificantly increased in human AD brains (Fig. 5, B and C). In addition,glypican-3 in the GDN fraction was elevated (P < 0.05) in AD brains,although there was no change in the TBSX fraction (Fig. 5D). Glypican-1, an abundant glypican in neurons, was trending to an increase in theGDN fractions, although this was not significant (Fig. 5E). We nextexamined the amounts of secreted extracellular matrix HSPGs, includ-ing agrin and perlecan, in these human brain tissues and found that theamount of agrin was increased in the GDN fractions of AD brains (Fig.5F; P < 0.01). A previous study showed that a large fraction of the agrinin AD brains is detergent-insoluble (49). In our measurements, theamount of agrin in the TBSX fractions was under the detection limitusing currently available ELISAs. Finally, the amount of perlecan waselevated in both detergent-soluble (P < 0.01) and detergent-insolublefractions (P < 0.05) in AD brain tissues (Fig. 5G). Together, these findingsdemonstrate that a number of HSPGs, distributed abundantly in the in-soluble compartments, were significantly increased in human AD brains.

DISCUSSION

The accumulation and deposition of Ab in the brain have been hy-pothesized to drive the pathogenic cascades of AD (1, 2). Substantialamyloid buildup is present in AD brains long before the clinical onset

ceTranslationalMedicine.org

by guest on March 26, 2021

iencemag.org/

of the disease (50, 51). Mounting studieshave clearly demonstrated the causes ofAb accumulation in early-onset familialAD patients as autosomal dominant genemutations lead to increased Ab42 pro-duction (52, 53). However, most AD casesare sporadic and late-onset, and the path-ological mechanisms that lead to Ab ac-cumulation are less clear. HSPGs areglycoproteins composed of a core proteincoupled to one or more HS glycosamino-glycan chains (12). Five distinct classes ofHSPGs have been identified, includingmembrane HSPGs (such as syndecansand glypicans) and the secreted extra-cellular matrix HSPGs (agrin, perlecan,and type XVIII collagen) (54). HSPGshave been shown to interact with a vari-ety of molecules because of their highlysulfated nature, thereby mediatingbiological activities including endocytosisand cell signaling (12, 55, 56). Here, wehave provided in vivo evidence that neu-ronal depletion of HS leads to reducedAb and amyloid deposition in APP/PS1amyloid model mice. Deficiency of neu-ronal HS did not affect APP processingand Ab production but did enhance Abclearance. These results suggest that theassociation of Ab and HSPGs on the sur-face of neurons likely suppresses or rep-resents an inefficient pathway for Ab

Fig. 4. Deletion of neuronal HS increases the abundance of CAA along the cerebral vasculature.(A ) Ab deposition in brain sections from control APP/PS1 and APP/PS1; nExt1CKOmice at 12 months of age

was immunostained with a pan-Ab antibody. Scale bar, 200 mm. (Inset) Immunostaining of Ab depositionalong leptomeningeal arteries as CAA in control APP/PS1 and APP/PS1; nExt1CKO mice. Scale bar, 10 mm.(B) The burden of CAA formation in leptomeningeal arteries in control APP/PS1 and APP/PS1; nExt1CKO

mice (10 arteries per mouse; 10 mice per genotype) was quantified after scanning Ab immunostaining bythe Positive Pixel Count program (Aperio Technologies). Data represent means ± SEM. **P < 0.01. Statis-tical analysis was performed using Student’s t test. (C) Coimmunofluorescence staining of leptomeningealarteries with aSMA (green) and Ab (red) antibodies. Scale bar, 10 mm.

30 March 2016 Vol 8 Issue 332 332ra44 6

R E S EARCH ART I C L E

by guest on March 26, 2021

http://stm.sciencem

ag.org/D

ownloaded from

clearance (fig. S7). Also, neuronal HSPGs might facilitate Ab oligo-merization and aggregation (fig. S7). Ab40 was shown to inhibit theheparanase-mediated degradation of HS (31, 57). Thus, the increasesof Ab and HS may result in a vicious cycle that further contributes tothe persistence and stability of amyloid plaques in AD brains.

Ab is mainly generated in neurons and eliminated from the brainthrough several clearance pathways (6, 46, 58–60); one of the majorpathways depends on enzymatic degradation in ISF by proteases such asneprilysin and insulin-degrading enzyme (61). Thus, the interactions of Abwith other molecules are predicted to interfere with the enzymatic deg-radation machineries by physically blocking their association sites.HSPGs have been proposed to function as a chaperone that serves as aprotective shield against the proteolytic degradation ofAb; thus, the clear-ance ofAb is likely hindered in thepresence of cell surfaceHSPGs (fig. S7)(47, 62). Consistent with this notion, our results demonstrated that defi-ciency of neuronal HS accelerates Ab elimination in themouse hippocam-pus.We found that deletion ofHS in neurons exacerbatedCAA formationdespite the reduction of Ab and amyloid plaque deposition in the brainparenchyma. Given that significant portions of ISF Ab flow along thecerebral vasculature and are cleared into theblood flow through the blood-brain barrier (BBB) or the perivascular drainage pathway (63, 64), Ab likely

www.Scien

is deposited on cerebrovasculature as CAA when these pathways aredisturbed or the amount of drained Ab overwhelms the clearance ca-pacity.Our results suggest that the expression ofHSPGs in various braincompartments might play an important role in regulating thedistribution of Ab in the brain. HSPGs on the surface of the neuronslikely capture Ab and serve as a reservoir for Ab accumulation in thebrain parenchyma.When this pathway is absent, the drainage of ISFAbinto the perivascular regions is likely enhanced. Specific HSPGs havebeen shown to be prominent within the cerebral microvasculatureand are involved in BBB maintenance, permeability, and the patho-genesis of CAA (31, 49). It is possible that the remaining HSPGs in themicrovascular network of the brain contribute to the deposition of cere-brovascular amyloid in the APP/PS1; nExt1CKO mice. Our neuronalHS-deficient mouse model provides further insight into the dynamicdistribution of Ab, and the potential mechanisms by which HSPGs indifferent brain compartments might mediate Ab metabolism in vivo.Whether vascular HSPGs directly affect BBB integrity, the perivasculardrainage clearance of Ab and CAA formation in vivo warrants furtherinvestigation.

HSPGs have also been suggested to function as a catalyst to promoteAb oligomer formation and aggregation (23, 62, 65). In vitro experiments

Fig. 5. Several classes of HSPGs are elevated in human AD brain tis-sues. Human temporal lobe brain tissues from control (Ctrl) (n = 20) and

E) The amounts of the GPI-anchored HSPGs glypican-1 and glypican-3 inTBSX and GDN fractions of control and AD brain tissue were quantified by

AD (n = 20) groups were lysed and fractionated through sequential extrac-tion with TBS, TBSX, and GDN. (A to C) Transmembrane HSPGs, includingsyndecan-1, syndecan-3, and syndecan-4, in TBSX and GDN fractions ofcontrol and AD brain tissues were quantified by specific ELISAs. (D and

specific ELISAs. (F and G) The amounts of two major extracellular matrixHSPGs agrin and perlecan in GDN fractions of control and AD brains werequantified by specific ELISAs. Results are presented as means ± SEM. *P <0.05; **P < 0.01.

ceTranslationalMedicine.org 30 March 2016 Vol 8 Issue 332 332ra44 7

R E S EARCH ART I C L E

by guest on March 26, 2021

http://stm.sciencem

ag.org/D

ownloaded from

have shown that HS accelerates Ab aggregation through their directinteraction (20–22). Membrane-bound glypican-1 has been shown to in-teractwith oligomerizedAb in the detergent-insoluble glycosphingolipid-enriched compartment (20). Agrin, an extracellular matrix HSPG, hasbeen found to bind to Ab and accelerated its fibril formation in humanAD brain (16). Indeed, several HSPGs colocalize with senile plaques inAD brains (31, 36). We also demonstrated that several subtypes ofHSPGs, including syndecan-3, syndecan-4, glypican-1, and glypican-3,were significantly higher in AD human brain tissues, particularly in thepools of detergent-insoluble fractions compared to those in control indi-viduals. These results suggest that aberrant amounts of HSPGs and theiraltered distribution inADbrainsmight contribute to Ab aggregation andplaque formation, although further studies are needed. It is also possiblethat amyloid plaques capture HSPGs or promote the accumulation ofHSPG-expressing glial cells, resulting in an increase of HSPGs in ADbrains (32).

Receptor-mediated cellular uptake and subsequent lysosomal deg-radation are other critical pathways for Ab clearance (64). Whereasmicroglia and astrocytes actively internalize and clear Ab, neurons alsoplay a role in cellular Ab clearance. Because HSPG is one of the majorplayers that capture Ab on the cell surface of neurons for endocytosis(64, 66), the deficiency of HS in neurons seems to be disadvantageousfor this clearance pathway. However, when excess amounts of Ab areinternalized, it can aggregate in lysosomes, likely providing “seeds” toinitiate further Ab aggregation (67, 68). In addition, Ab has beenshown to be self-propagating and is transmitted from one neuronto another after its cellular uptake (69–71). Thus, although the neuronaluptake of Ab through HSPGs can lead to trafficking to the lysosomes(25), albeit less efficiently than via the low-density lipoprotein receptor–related protein 1 (LRP1) (66), such a pathway might be harmful whenthe capacity for lysosomal degradation of Ab is overwhelmed and com-promised (fig. S7). It will be critical to investigate whether neuronalHSPGs mediate the propagation of Ab during AD pathogenesis in fu-ture studies. Extracellular tau and a-synuclein have been shown topropagate after being endocytosed via cell surface HSPGs on neurons(56), which is likely to be a critical step for the progression of relatedneurodegenerative diseases (72). Thus, blocking the association of thesepathogenic molecules with neuronal HSPGs may ameliorate the pro-gression of various neurodegenerative diseases by preventing their prop-agation between neurons during disease development.

HSPGs interact with myriad molecules that orchestrate importantbiological functions (13). One limitation of our study is that we cannotrule out the possibility that disruption ofHS binding to othermolecules,in addition to Ab, may also contribute to the modulation of Abmetab-olism in the adult mouse brain. One such molecule is apolipoprotein E(apoE), whose binding to HSPGs is known to contribute to the metab-olism of apoE itself (64, 73). The e4 allele of theAPOE gene is the stron-gest genetic risk factor for sporadic AD among the three polymorphicalleles (e2, e3, and e4). ApoE is required for seeding amyloid and hasbeen shown to regulate Ab metabolism and amyloid deposition in anapoE isoform–dependent manner (6, 74). Whether apoE is involved inHSPG-regulated Ab metabolism/aggregation and whether this effectdepends on apoE isoforms require further investigation.

In summary, our studies using genetically altered mice have clearlydemonstrated a critical role of neuronal HS in brain Ab clearance andoligomerization. Given that neuronal HSPGs are also likely involved inthe pathogenic propagation of various proteins in neurodegenerativediseases and HSPGs are elevated in human brain tissues from AD pa-

www.Scien

tients, our studies establish a rationale for targeting the Ab-HSPGpathway for therapy. Toward this end, it is noteworthy that the admin-istration of a low–molecular weight heparin reduced amyloid pathologyin a mouse model of AD, likely by antagonizing Ab-HSPG interactions(65). Given that HSPGs interact withmyriad critical molecules thatme-diate diverse biological activities (55), it will be critical to design andidentify compounds that specifically inhibit the interaction betweenAb and HSPGs for the treatment of neurodegenerative diseases. Whencombined with other strategies targeting Ab, including immuno-therapy, these approaches might allow for a reduction, if not elimina-tion, of Ab-related toxicity.

MATERIALS AND METHODS

Study designThis study aimed to investigate the role of HSPGs in regulating brainAb metabolism using a conditional knockout mouse model. To ac-complish this aim, the Ext1 gene, which encodes an essential glycosyl-transferase for HS biosynthesis, was genetically ablated in theforebrain neurons of APP/PS1 mice. The effects of neuronal HS defi-ciency on the amount of Ab, amyloid deposition, and neuroinflamma-tion were assessed by Ab ELISAs, immunohistochemistry, real-timePCR, and Western blot analyses. The neuronal HS-deficient APP/PS1 mice exhibited marked reduction in amyloid oligomerizationand deposition. Using in vivo microdialysis, we also detected anaccelerated rate of Ab clearance in the brain ISF of HS-deficient mice,suggesting that neuronal HS might represent an inefficient or inhibi-tory pathway for Ab clearance. We further examined the amounts ofvarious HSPG species in postmortem human brain tissues from con-trol and AD patients using specific ELISAs to evaluate the potentialrelevance of altered HSPGs in AD pathogenesis. The mice of differentgenotypes were selected on the basis of availability. Sample sizes wereadequately powered to observe the effects on the basis of past experi-ence of animal studies (66, 75). Data collection and the quantificationof immunoreactivity for mouse samples were performed with the in-vestigators unaware of the sample identities until statistical analyses.

Human postmortem brain tissuesTemporal lobe cortical samples from neurologically unimpairedsubjects (n = 20) and from subjects with AD (n = 20) were obtainedfrom the University of Kentucky Alzheimer’s Disease Center (ADC)Neuropathology Core (Lexington, KY) (75). After autopsy, the sampleswere snap-frozen in liquid nitrogen and kept frozen at −80°C. Informedconsent was obtained through ADCNeuropathology Core, and we ob-tained institutional review board approval to use these brain samples forresearch. Diagnosis of AD was confirmed by pathological and clinicalcriteria as described (76). The average age of subjects was 85.1 ± 5.7 yearsin the control group and 84.4 ± 5.2 years in the AD group (P = 0.69).Average postmortem interval was 3.2 hours andwas not significantly dif-ferent between the two groups (P = 0.22).

AnimalsAll animal procedures were approved by the Mayo Clinic InstitutionalAnimal Care and Use Committee and were in accordance with theNational Institutes of Health Guide for the Care and Use of LaboratoryAnimals. Ext1flox/flox mice have been described in previous studies (27).Double transgenic APPswe/PS1DE9 (APP/PS1)mice were purchased from

ceTranslationalMedicine.org 30 March 2016 Vol 8 Issue 332 332ra44 8

R E S EARCH ART I C L E

by guest on March 26, 2021

http://stm.sciencem

ag.org/D

ownloaded from

The Jackson Laboratory and were initially generated by Jankowsky et al.as described (29).

In vivo Ab microdialysisTo assess ISF Ab in the hippocampus of awake, freely moving APP/PS1 and APP/PS1; nExt1CKO mice, in vivo microdialysis was per-formed as previously described (8, 44). Briefly, under isoflurane vola-tile anesthetic, guide cannula (BR style; Bioanalytical Systems) wascemented into the hippocampus (3.1 mm behind bregma, 2.5 mm lat-eral to midline, and 1.2 mm below dura at a 12° angle). A micro-dialysis probe (38-kD molecular weight cutoff membrane; BioanalyticalSystems) was inserted through the guide cannula into the brain. Artifi-cial CSF (1.3 mM CaCl2, 1.2 mM MgSO4, 3 mM KCl, 0.4 mMKH2PO4, 25 mM NaHCO3, and 122 mM NaCl, pH 7.35) containing4% bovine serum albumin (BSA) (Sigma) filtered through a 0.1-mmmembrane was used as microdialysis perfusion buffer. Flow ratewas a constant 1.0 ml/min. Samples were collected every 60 to 90 minovernight, which gets through the 4- to 6-hour recovery period, andthe mean concentration of Ab over the 6-hour preceding treatmentwas defined as basal concentration of ISF Ab. Samples were collectedthrough a refrigerated fraction collector and assessed for Ab40 orAb42 by ELISAs. For each animal, all Ab concentrations were normal-ized to the basal Ab concentration. To measure Ab elimination half-life, mice were administered a g-secretase inhibitor, LY411575 (5 mg/kg),intraperitoneally to rapidly block the production of Ab. Microdialysissamples were collected every 60 min for 6 hours. The half-life of ISFAb was calculated on the basis of the slope of the semi-log plot of per-cent change in Ab versus time (8).

ELISA quantificationISF Ab concentrations were assessed using sandwich Ab40- or Ab42-specific ELISAs as described (44). Briefly, a mouse anti-Ab40 antibodyHJ2 (anti–Ab35-40) or Ab42 antibody HJ7.4 (anti–Ab37-42) was usedas capture antibody, respectively. A biotinylated antibody HJ5.1 (anti–Ab13-28) targeting the central domain of Ab was used as the detectingantibody, followed by ELISA development using streptavidin poly-HRP40 (Fitzgerald Industries). Synthetic human Ab40 or Ab42 pep-tide (American Peptide) was used to generate the standard curves foreach assay. The ELISA assays were developed using Super Slow ELISATMB (Sigma) with absorbance read on a BioTek plate reader. The Abconcentration in the brain lysates was determined by ELISA (77) withend-specific monoclonal antibody (mAb) 2.1.3 (human Abx–42–specific)and mAb 13.1.1 (human Abx–40–specific) for capture and horseradishperoxidase (HRP)–conjugated mAb Ab5 (human Ab1–16–specific) fordetection. The ELISAs were developed using Super Slow ELISA TMB(Sigma). To detect soluble oligomeric Ab species, the same antibody,3D6 (to Ab residues 1 to 5), was used for both capture and detectionsimilarly as previously described (40). Briefly, 3D6 (10 mg/ml) antibodywas coated into 96-well immunoassay plates overnight at 4°C. Theplates were then aspirated and blocked with 4% BSA in phosphate-buffered saline (PBS) buffer for at least 1 hour at 37°C. The samplesand standards were added to the plates and incubated overnight at 4°C.The Ab oligomers used as standards were prepared and characterized aspreviously described (78, 79). The plates were washed three or more timeswith wash buffer between each step of the assay. The plate was incu-bated with the biotinylated 3D6 (0.5 mg/ml) antibody in assay buffer(1% BSA, PBS) for 90 min at 37°C. The avidin-HRP (Vector Labora-tories) was added to the wells for 90 min at room temperature. The

www.Scien

ELISAs were developed using Super Slow ELISA TMB (Sigma). Tomeasure total Ab amount, 3D6 (10 mg/ml) and biotinylated mAb266 (to Ab residues 13 to 28, 0.5 mg/ml) were used as the captureand detection antibodies, respectively. All procedures were the sameas those in oligomeric Ab ELISA described above.

Preparation of brain homogenatesMouse brain tissues and human postmortem brain tissues were pro-cessed through sequential extraction as described (38). Frozen brain tis-sues were homogenized with TBS and centrifuged at 100,000g for 60minat 4°C with supernatant defined as TBS-soluble fraction. Pellets wereresuspended in TBS buffer containing 1% Triton X-100 (TBSX) andmixed gently by rotation at 4°C for 30 min, followed by a second cen-trifugation at 100,000g for 60 min with supernatant defined as TBSX-soluble fraction. The TBSX-insoluble pellet was resuspended with 5 Mguanidine, mixed by rotation at room temperature overnight, and cen-trifuged at 16,000g for 30minwith supernatant defined asGDN-solublefraction.

Immunohistochemical and immunofluorescence stainingand analysesParaffin-embedded sections were immunostained using pan-Ab (Ab33.1.1; human Ab1–16–specific), anti-GFAP (BioGenex), anti-Ab40(Millipore), anti-Ab42 (21F12, provided by E. Lilly), and anti-Iba1(Wako) antibodies (80). Immunohistochemically stained sections werecaptured using the ImageScope AT2 image scanner (Aperio Technol-ogies) and analyzed using the ImageScope software. The intensities ofGFAP and Iba1 staining in the hippocampus were calculated using thePositive Pixel Count program available with the ImageScope software(Aperio Technologies) (80). The fibrillar Ab was immunostained withthioflavin S. The images were captured by Aperio Fluorescent Scanner,and the stained areas covered by the fibrillary plaque were quantifiedby ImageJ. For examining the CAA in the cerebral vasculature, paraffin-embedded sections were costained with aSMA antibody (Abcam)and Ab antibody (MOAB2), followed by Alexa Fluor 488– and AlexaFluor 568–conjugated secondary antibodies (Invitrogen). MOAB2antibody was a gift fromM. J. LaDu (University of Illinois at Chicago).The images were acquired by a confocal laser scanning fluorescencemicroscope (model LSM 510 invert, Carl Zeiss).

Statistical analysisAll quantified data represent an average of samples. Statistical analyseswere performed with Excel or GraphPad Prism software. Statistical sig-nificance was determined by two-tailed Student’s t test. Levels of signif-icance are as follows: *P < 0.05, **P < 0.01; P < 0.05 was consideredsignificant.

SUPPLEMENTARY MATERIALS

www.sciencetranslationalmedicine.org/cgi/content/full/8/332/332ra44/DC1Materials and MethodsFig. S1. Characterization of neuronal HS-deficient mice.Fig. S2. The mRNA expression of HSPG subfamily in APP/PS1 and APP/PS1; nExt1CKO mice.Fig. S3. HSPG core proteins codeposit with amyloid plaques in the brain parenchyma andleptomeningeal arteries.Fig. S4. Deficiency of neuronal HS leads to a decrease in amyloid-associated astrogliosis.Fig. S5. Deficiency of neuronal HS increases the clearance of ISF Ab42 in the hippocampus ofAPP/PS1 mice.

ceTranslationalMedicine.org 30 March 2016 Vol 8 Issue 332 332ra44 9

R E S EARCH ART I C L E

Fig. S6. Ab40 and Ab42 deposition is reduced in brain parenchyma, but enhanced along thecerebral vasculature in neuronal HS-deficient mice.Fig. S7. Proposed mechanism by which neuronal HS/HSPGs may regulate brain Ab clearanceand aggregation.Table S1. Demographic characteristics of examined controls and individuals with AD.Table S2. Ab concentration in the cortex and hippocampus in APP/PS1 and APP/PS1; nExt1CKOmice.

by guest on March 26, 2021

http://stm.sciencem

ag.org/D

ownloaded from

REFERENCES AND NOTES

1. J. Hardy, D. J. Selkoe, The amyloid hypothesis of Alzheimer’s disease: Progress and problemson the road to therapeutics. Science 297, 353–356 (2002).

2. D. J. Selkoe, Deciphering the genesis and fate of amyloid b-protein yields novel therapiesfor Alzheimer disease. J. Clin. Invest. 110, 1375–1381 (2002).

3. R. Vassar, B. D. Bennett, S. Babu-Khan, S. Kahn, E. A. Mendiaz, P. Denis, D. B. Teplow, S. Ross,P. Amarante, R. Loeloff, Y. Luo, S. Fisher, J. Fuller, S. Edenson, J. Lile, M. A. Jarosinski, A. L. Biere,E. Curran, T. Burgess, J.-C. Louis, F. Collins, J. Treanor, G. Rogers, M. Citron, b-Secretase cleavageof Alzheimer’s amyloid precursor protein by the transmembrane aspartic protease BACE.Science 286, 735–741 (1999).

4. W. T. Kimberly, M. J. LaVoie, B. L. Ostaszewski, W. Ye, M. S. Wolfe, D. J. Selkoe, g-Secretase isa membrane protein complex comprised of presenilin, nicastrin, aph-1, and pen-2. Proc. Natl.Acad. Sci. U.S.A. 100, 6382–6387 (2003).

5. A. Goate, Segregation of a missense mutation in the amyloid b-protein precursor genewith familial Alzheimer’s disease. J. Alzheimers Dis. 9, 341–347 (2006).

6. G. Bu, Apolipoprotein E and its receptors in Alzheimer’s disease: Pathways, pathogenesisand therapy. Nat. Rev. Neurosci. 10, 333–344 (2009).

7. L. Mucke, D. J. Selkoe, Neurotoxicity of amyloid b-protein: Synaptic and network dys-function. Cold Spring Harb. Perspect. Med. 2, a006338 (2012).

8. J. R. Cirrito, P. C.May,M.A.O’Dell, J.W. Taylor,M.Parsadanian, J.W. Cramer, J. E. Audia, J. S. Nissen,K. R. Bales, S. M. Paul, R. B. DeMattos, D. M. Holtzman, In vivo assessment of brain interstitial fluidwithmicrodialysis reveals plaque-associated changes in amyloid-bmetabolism and half-life.J. Neurosci. 23, 8844–8853 (2003).

9. R. J. Bateman, L. Y. Munsell, J. C. Morris, R. Swarm, K. E. Yarasheski, D. M. Holtzman, Humanamyloid-b synthesis and clearance rates as measured in cerebrospinal fluid in vivo. Nat.Med. 12, 856–861 (2006).

10. K. G. Mawuenyega, W. Sigurdson, V. Ovod, L. Munsell, T. Kasten, J. C. Morris, K. E. Yarasheski,R. J. Bateman, Decreased clearance of CNS b-amyloid in Alzheimer’s disease. Science 330,1774 (2010).

11. P. W. Park, O. Reizes, M. Bernfield, Cell surface heparan sulfate proteoglycans: Selectiveregulators of ligand-receptor encounters. J. Biol. Chem. 275, 29923–29926 (2000).

12. S. Sarrazin, W. C. Lamanna, J. D. Esko, Heparan sulfate proteoglycans. Cold Spring Harb.Perspect. Biol. 3, a004952 (2011).

13. J. R. Bishop, M. Schuksz, J. D. Esko, Heparan sulphate proteoglycans fine-tune mammalianphysiology. Nature 446, 1030–1037 (2007).

14. T. Lind, F. Tufaro, C. McCormick, U. Lindahl, K. Lidholt, The putative tumor suppressorsEXT1 and EXT2 are glycosyltransferases required for the biosynthesis of heparan sulfate.J. Biol. Chem. 273, 26265–26268 (1998).

15. X. Lin, G. Wei, Z. Shi, L. Dryer, J. D. Esko, D. E. Wells, M. M. Matzuk, Disruption of gastrulationand heparan sulfate biosynthesis in EXT1-deficient mice. Dev. Biol. 224, 299–311 (2000).

16. S. L. Cotman, W. Halfter, G. J. Cole, Agrin binds to b-amyloid (Ab), accelerates Ab fibrilformation, and is localized to Ab deposits in Alzheimer’s disease brain. Mol. Cell. Neurosci.15, 183–198 (2000).

17. J. van Horssen, I. Otte-Höller, G. David, M. L. Maat-Schieman, L. P. van den Heuvel, P. Wesseling,R. M. de Waal, M. M. Verbeek, Heparan sulfate proteoglycan expression in cerebrovascularamyloid b deposits in Alzheimer’s disease and hereditary cerebral hemorrhage with amyloi-dosis (Dutch) brains. Acta Neuropathol. 102, 604–614 (2001).

18. D. Van Gool, G. David, M. Lammens, F. Baro, R. Dom, Heparan sulfate expression patterns inthe amyloid deposits of patients with Alzheimer’s and Lewy body type dementia. Dementia 4,308–314 (1993).

19. G.-l. Zhang, X. Zhang, X.-m. Wang, J.-P. Li, Towards understanding the roles of heparansulfate proteoglycans in Alzheimer’s disease. Biomed. Res. Int. 2014, 516028 (2014).

20. N. Watanabe, W. Araki, D.-H. Chui, T. Makifuchi, Y. Ihara, T. Tabira, Glypican-1 as an Abbinding HSPG in the human brain: Its localization in DIG domains and possible roles inthe pathogenesis of Alzheimer’s disease. FASEB J. 18, 1013–1015 (2004).

21. F. Cheng, K. Ruscher, L.-Å. Fransson, K. Mani, Non-toxic amyloid beta formed in the presence ofglypican-1 or its deaminatively generated heparan sulfate degradation products. Glycobiology23, 1510–1519 (2013).

22. K. R. Brunden, N. J. Richter-Cook, N. Chaturvedi, R. C. A. Frederickson, pH-dependentbinding of synthetic b-amyloid peptides to glycosaminoglycans. J. Neurochem. 61,2147–2154 (1993).

www.Scienc

23. E. Sandwall, P. O’Callaghan, X. Zhang, U. Lindahl, L. Lannfelt, J.-P. Li, Heparan sulfate mediatesamyloid-beta internalization and cytotoxicity. Glycobiology 20, 533–541 (2010).

24. L. Bergamaschini, C. Donarini, E. Rossi, A. De Luigi, C. Vergani, M. G. De Simoni, Heparinattenuates cytotoxic and inflammatory activity of Alzheimer amyloid-b in vitro. Neurobiol.Aging 23, 531–536 (2002).

25. T. Kanekiyo, J. Zhang, Q. Liu, C.-C. Liu, L. Zhang, G. Bu, Heparan sulphate proteoglycan andthe low-density lipoprotein receptor-related protein 1 constitute major pathways for neu-ronal amyloid-b uptake. J. Neurosci. 31, 1644–1651 (2011).

26. C. B. Jendresen, H. Cui, X. Zhang, I. Vlodavsky, L. N. G. Nilsson, J.-P. Li, Overexpression ofheparanase lowers the amyloid burden in amyloid-b precursor protein transgenic mice. J. Biol.Chem. 290, 5053–5064 (2015).

27. M. Inatani, F. Irie, A. S. Plump, M. Tessier-Lavigne, Y. Yamaguchi, Mammalian brain mor-phogenesis and midline axon guidance require heparan sulfate. Science 302, 1044–1046(2003).

28. J. Z. Tsien, D. F. Chen, D. Gerber, C. Tom, E. H. Mercer, D. J. Anderson, M. Mayford, E. R. Kandel,S. Tonegawa, Subregion- and cell type–restricted gene knockout in mouse brain. Cell 87,1317–1326 (1996).

29. J. L. Jankowsky, D. J. Fadale, J. Anderson, G. M. Xu, V. Gonzales, N. A. Jenkins, N. G. Copeland,M. K. Lee, L. H. Younkin, S. L. Wagner, S. G. Younkin, D. R. Borchelt, Mutant presenilins specif-ically elevate the levels of the 42 residue b-amyloid peptide in vivo: Evidence for augmenta-tion of a 42-specific g secretase. Hum. Mol. Genet. 13, 159–170 (2004).

30. F. Irie, H. Badie-Mahdavi, Y. Yamaguchi, Autism-like socio-communicative deficits andstereotypies in mice lacking heparan sulfate. Proc. Natl. Acad. Sci. U.S.A. 109, 5052–5056(2012).

31. J. van Horssen, P. Wesseling, L. P. W. J. van den Heuvel, R. M. W. de Waal, M. M. Verbeek,Heparan sulphate proteoglycans in Alzheimer’s disease and amyloid-related disorders.Lancet Neurol. 2, 482–492 (2003).

32. P. O’Callaghan, E. Sandwall, J.-P. Li, H. Yu, R. Ravid, Z.-Z. Guan, T. H. VanKuppevelt, L. N. G. Nilsson,M. Ingelsson, B. T. Hyman, H. Kalimo, U. Lindahl, L. Lannfelt, X. Zhang, Heparan sulfate ac-cumulation with Ab deposits in Alzheimer’s disease and Tg2576 mice is contributed by glialcells. Brain Pathol. 18, 548–561 (2008).

33. M. Garcia-Alloza, E. M. Robbins, S. X. Zhang-Nunes, S. M. Purcell, R. A. Betensky, S. Raju,C. Prada, S. M. Greenberg, B. J. Bacskai, M. P. Frosch, Characterization of amyloid deposi-tion in the APPswe/PS1dE9 mouse model of Alzheimer disease. Neurobiol. Dis. 24, 516–524(2006).

34. K. L. Youmans, L. M. Tai, T. Kanekiyo, W. B. Stine Jr., S.-C.Michon, E. Nwabuisi-Heath, A.M.Manelli,Y. Fu, S. Riordan, W. A. Eimer, L. Binder, G. Bu, C. Yu, D. M. Hartley, M. J. LaDu, Intraneuronal Abdetection in 5xFAD mice by a new Ab-specific antibody. Mol. Neurodegener. 7, 8 (2012).

35. N. M. Timmer, M. K. Herbert, J. W. Kleinovink, A. J. Kiliaan, R. M. W. De Waal, M. M. Verbeek,Limited expression of heparan sulphate proteoglycans associated with Ab deposits in theAPPswe/PS1dE9 mouse model for Alzheimer’s disease. Neuropathol. Appl. Neurobiol. 36,478–486 (2010).

36. M. M. Verbeek, I. Otte-Höller, J. van den Born, L. P. W. J. van denHeuvel, G. David, P. Wesseling,R. M. W. de Waal, Agrin is a major heparan sulfate proteoglycan accumulating in Alzheimer’sdisease brain. Am. J. Pathol. 155, 2115–2125 (1999).

37. S. M. Rauch, K. Huen, M. C. Miller, H. Chaudry, M. Lau, J. R. Sanes, C. E. Johanson, E. G. Stopa,R. W. Burgess, Changes in brain b-amyloid deposition and aquaporin 4 levels in responseto altered Agrin expression in mice. J. Neuropathol. Exp. Neurol. 70, 1124–1137 (2011).

38. K. L. Youmans, S. Leung, J. Zhang, E. Maus, K. Baysac, G. Bu, R. Vassar, C. Yu, M. J. LaDu,Amyloid-b42 alters apolipoprotein E solubility in brains of mice with five familial AD mutations.J. Neurosci. Methods 196, 51–59 (2011).

39. D. J. Selkoe, Soluble oligomers of the amyloid b-protein impair synaptic plasticity and be-havior. Behav. Brain Res. 192, 106–113 (2008).

40. W. Xia, T. Yang, G. Shankar, I. M. Smith, Y. Shen, D. M. Walsh, D. J. Selkoe, A specific enzyme-linked immunosorbent assay for measuring b-amyloid protein oligomers in human plasmaand brain tissue of patients with Alzheimer disease. Arch. Neurol. 66, 190–199 (2009).

41. G. M. Castillo, W. Lukito, T. N. Wight, A. D. Snow, The sulfate moieties of glycosaminoglycansare critical for theenhancement ofb-amyloidprotein fibril formation. J. Neurochem.72, 1681–1687(1999).

42. T. Wyss-Coray, Inflammation in Alzheimer disease: Driving force, bystander or beneficialresponse? Nat. Med. 12, 1005–1015 (2006).

43. J. M. Rubio-Perez, J. M. Morillas-Ruiz, A review: Inflammatory process in Alzheimer’s disease,role of cytokines. ScientificWorldJournal 2012, 756357 (2012).

44. J. R. Cirrito, B. M. Disabato, J. L. Restivo, D. K. Verges, W. D. Goebel, A. Sathyan, D. Hayreh,G. D’Angelo, T. Benzinger, H. Yoon, J. Kim, J. C. Morris, M. A. Mintun, Y. I. Sheline, Serotoninsignaling is associated with lower amyloid-b levels and plaques in transgenic mice andhumans. Proc. Natl. Acad. Sci. U.S.A. 108, 14968–14973 (2011).

45. A. W. Bero, P. Yan, J. H. Roh, J. R. Cirrito, F. R. Stewart, M. E. Raichle, J.-M. Lee, D. M. Holtzman,Neuronal activity regulates the regional vulnerability to amyloid-b deposition. Nat. Neurosci.14, 750–756 (2011).

eTranslationalMedicine.org 30 March 2016 Vol 8 Issue 332 332ra44 10

R E S EARCH ART I C L E

by guest on March 26, 2021

http://stm.sciencem

ag.org/D

ownloaded from

46. R. O. Weller, M. Subash, S. D. Preston, I. Mazanti, R. O. Carare, Perivascular drainage ofamyloid-b peptides from the brain and its failure in cerebral amyloid angiopathy andAlzheimer’s disease. Brain Pathol. 18, 253–266 (2008).

47. A. D. Snow, R. T. Sekiguchi, D. Nochlin, R. N. Kalaria, K. Kimata, Heparan sulfate proteo-glycan in diffuse plaques of hippocampus but not of cerebellum in Alzheimer’s diseasebrain. Am. J. Pathol. 144, 337–347 (1994).

48. A. D. Snow, H. Mar, D. Nochlin, K. Kimata, M. Kato, S. Suzuki, J. Hassell, T. N. Wight, Thepresence of heparan sulfate proteoglycans in the neuritic plaques and congophilic angi-opathy in Alzheimer’s disease. Am. J. Pathol. 133, 456–463 (1988).

49. J. E. Donahue, T. M. Berzin, M. S. Rafii, D. J. Glass, G. D. Yancopoulos, J. R. Fallon, E. G. Stopa,Agrin in Alzheimer’s disease: Altered solubility and abnormal distribution within micro-vasculature and brain parenchyma. Proc. Natl. Acad. Sci. U.S.A. 96, 6468–6472 (1999).

50. Y. Huang, L. Mucke, Alzheimer mechanisms and therapeutic strategies. Cell 148, 1204–1222(2012).

51. D. M. Holtzman, J. C. Morris, A. M. Goate, Alzheimer’s disease: The challenge of the secondcentury. Sci. Transl. Med. 3, 77sr71 (2011).

52. M. Citron, T. Oltersdorf, C. Haass, L. McConlogue, A. Y. Hung, P. Seubert, C. Vigo-Pelfrey,I. Lieberburg, D. J. Selkoe, Mutation of the b-amyloid precursor protein in familial Alzheimer’sdisease increases b-protein production. Nature 360, 672–674 (1992).

53. D. Campion, C. Dumanchin, D. Hannequin, B. Dubois, S. Belliard, M. Puel, C. Thomas-Anterion,A. Michon, C. Martin, F. Charbonnier, G. Raux, A. Camuzat, C. Penet, V. Mesnage, M. Martinez,F. Clerget-Darpoux, A. Brice, T. Frebourg, Early-onset autosomal dominant Alzheimer disease:Prevalence, genetic heterogeneity, and mutation spectrum. Am. J. Hum. Genet. 65, 664–670(1999).

54. R. V. Iozzo, Heparan sulfate proteoglycans: Intricate molecules with intriguing functions.J. Clin. Invest. 108, 165–167 (2001).

55. J. Turnbull, A. Powell, S. Guimond, Heparan sulfate: Decoding a dynamic multifunctionalcell regulator. Trends Cell Biol. 11, 75–82 (2001).

56. B. B. Holmes, S. L. DeVos, N. Kfoury, M. Li, R. Jacks, K. Yanamandra, M. O. Ouidja, F. M. Brodsky,J. Marasa, D. P. Bagchi, P. T. Kotzbauer, T. M. Miller, D. Papy-Garcia, M. I. Diamond, Heparansulfate proteoglycans mediate internalization and propagation of specific proteopathic seeds.Proc. Natl. Acad. Sci. U.S.A. 110, E3138–E3147 (2013).

57. K. J. Bame, J. Danda, A. Hassall, S. Tumova, Ab(1–40) prevents heparanase-catalyzed degrada-tion of heparan sulfate glycosaminoglycans and proteoglycans in vitro. A role for heparansulfate proteoglycan turnover in Alzheimers disease. J. Biol. Chem. 272, 17005–17011 (1997).

58. J. R. Cirrito, R. Deane, A. M. Fagan, M. L. Spinner, M. Parsadanian, M. B. Finn, H. Jiang, J. L. Prior,A. Sagare, K. R. Bales, S. M. Paul, B. V. Zlokovic, D. Piwnica-Worms, D.M.Holtzman, P-glycoproteindeficiency at the blood-brain barrier increases amyloid-b deposition in an Alzheimer diseasemouse model. J. Clin. Invest. 115, 3285–3290 (2005).

59. W. Farris, S. G. Schütz, J. R. Cirrito, G. M. Shankar, X. Sun, A. George, M. A. Leissring, D. M. Walsh,W. Q. Qiu, D. M. Holtzman, D. J. Selkoe, Loss of neprilysin function promotes amyloid plaqueformation and causes cerebral amyloid angiopathy. Am. J. Pathol. 171, 241–251 (2007).

60. J. S. Miners, S. Baig, J. Palmer, L. E. Palmer, P. G. Kehoe, S. Love, Ab-degrading enzymes inAlzheimer’s disease. Brain Pathol. 18, 240–252 (2008).

61. T. Saido, M. A. Leissring, Proteolytic degradation of amyloid b-protein. Cold Spring Harb.Perspect. Med. 2, a006379 (2012).

62. R. Gupta-Bansal, R. C. A. Frederickson, K. R. Brunden, Proteoglycan-mediated inhibition of Abproteolysis. A potential cause of senile plaque accumulation. J. Biol. Chem.270, 18666–18671 (1995).

63. A. P. Sagare, R. D. Bell, B. V. Zlokovic, Neurovascular dysfunction and faulty amyloid b-peptideclearance in Alzheimer disease. Cold Spring Harb. Perspect. Med. 2, a011452 (2012).

64. T. Kanekiyo, H. Xu, G. Bu, ApoE and Ab in Alzheimer’s disease: Accidental encounters orpartners? Neuron 81, 740–754 (2014).

65. L. Bergamaschini, E. Rossi, C. Storini, S. Pizzimenti, M. Distaso, C. Perego, A. De Luigi, C. Vergani,M. G. De Simoni, Peripheral treatment with enoxaparin, a low molecular weight heparin,reduces plaques and b-amyloid accumulation in a mouse model of Alzheimer’s disease.J. Neurosci. 24, 4181–4186 (2004).

66. T. Kanekiyo, J. R. Cirrito, C.-C. Liu, M. Shinohara, J. Li, D. R. Schuler, D. M. Holtzman, G. Bu,Neuronal clearance of amyloid-b by endocytic receptor LRP1. J. Neurosci. 33, 19276–19283 (2013).

67. X. Hu, S. L. Crick, G. Bu, C. Frieden, R. V. Pappu, J. M. Lee, Amyloid seeds formed by cellularuptake, concentration, and aggregation of the amyloid-beta peptide. Proc. Natl. Acad. Sci.U.S.A. 106, 20324–20329 (2009).

68. J. Li, T. Kanekiyo, M. Shinohara, Y. Zhang, M. J. LaDu, H. Xu, G. Bu, Differential regulation ofamyloid-b endocytic trafficking and lysosomal degradation by apolipoprotein E isoforms.J. Biol. Chem. 287, 44593–44601 (2012).

69. J. C. Watts, C. Condello, J. Stöhr, A. Oehler, J. Lee, S. J. DeArmond, L. Lannfelt, M. Ingelsson,K. Giles, S. B. Prusiner, Serial propagation of distinct strains of Ab prions from Alzheimer’sdisease patients. Proc. Natl. Acad. Sci. U.S.A. 111, 10323–10328 (2014).

www.Scienc

70. M. Meyer-Luehmann, J. Coomaraswamy, T. Bolmont, S. Kaeser, C. Schaefer, E. Kilger,A. Neuenschwander, D. Abramowski, P. Frey, A. L. Jaton, J.-M. Vigouret, P. Paganetti,D. M. Walsh, P. M. Mathews, J. Ghiso, M. Staufenbiel, L. C. Walker, M. Jucker, Exogenous in-duction of cerebral b-amyloidogenesis is governed by agent and host. Science 313, 1781–1784(2006).

71. H.-L. Song, S. Shim, D.-H. Kim, S.-H. Won, S. Joo, S. Kim, N. L. Jeon, S.-Y. Yoon, b-Amyloid istransmitted via neuronal connections along axonal membranes. Ann. Neurol. 75, 88–97(2014).

72. J. L. Guo, D. J. Covell, J. P. Daniels, M. Iba, A. Stieber, B. Zhang, D. M. Riddle, L. K. Kwong, Y. Xu,J. Q. Trojanowski, V. M. Y. Lee, Distinct a-synuclein strains differentially promote tau inclusionsin neurons. Cell 154, 103–117 (2013).

73. M. Futamura, P. Dhanasekaran, T. Handa, M. C. Phillips, S. Lund-Katz, H. Saito, Two-stepmechanism of binding of apolipoprotein E to heparin: Implications for the kinetics of apo-lipoprotein E-heparan sulfate proteoglycan complex formation on cell surfaces. J. Biol.Chem. 280, 5414–5422 (2005).

74. C.-C. Liu, T. Kanekiyo, H. Xu, G. Bu, Apolipoprotein E and Alzheimer disease: Risk, mechanismsand therapy. Nat. Rev. Neurol. 9, 106–118 (2013).

75. C.-C. Liu, C.-W. Tsai, F. Deak, J. Rogers, M. Penuliar, Y. M. Sung, J. N. Maher, Y. Fu, X. Li, H. Xu,S. Estus, H.-S. Hoe, J. D. Fryer, T. Kanekiyo, G. Bu, Deficiency in LRP6-mediated Wntsignaling contributes to synaptic abnormalities and amyloid pathology in Alzheimer’s dis-ease. Neuron 84, 63–77 (2014).

76. K. E. Grear, I.-F. Ling, J. F. Simpson, J. L. Furman, C. R. Simmons, S. L. Peterson, F. A. Schmitt,W. R. Markesbery, Q. Liu, J. E. Crook, S. G. Younkin, G. Bu, S. Estus, Expression of SORL1 anda novel SORL1 splice variant in normal and Alzheimers disease brain. Mol. Neurodegener. 4,46 (2009).

77. P. Das, C. Verbeeck, L. Minter, P. Chakrabarty, K. Felsenstein, T. Kukar, G. Maharvi, A. Fauq,B. A. Osborne, T. E. Golde, Transient pharmacologic lowering of Ab production prior todeposition results in sustained reduction of amyloid plaque pathology. Mol. Neurodegener.7, 39 (2012).

78. W. M. Tay, J. G. Bryant, P. K. Martin, A. J. Nix, B. M. Cusack, T. L. Rosenberry, A mass spec-trometric approach for characterization of amyloid-b aggregates and identification of theirpost-translational modifications. Biochemistry 51, 3759–3766 (2012).

79. V. Rangachari, B. D. Moore, D. K. Reed, L. K. Sonoda, A. W. Bridges, E. Conboy, D. Hartigan,T. L. Rosenberry, Amyloid-b(1–42) rapidly forms protofibrils and oligomers by distinct path-ways in low concentrations of sodium dodecylsulfate. Biochemistry 46, 12451–12462(2007).

80. P. Chakrabarty, C. Ceballos-Diaz, A. Beccard, C. Janus, D. Dickson, T. E. Golde, P. Das, IFN-gpromotes complement expression and attenuates amyloid plaque deposition in amyloid bprecursor protein transgenic mice. J. Immunol. 184, 5333–5343 (2010).

Acknowledgments: We thank S. Estus (University of Kentucky) and University of KentuckyADC (P30 AG028383) for providing postmortem human brain tissues, M. J. LaDu (Universityof Illinois at Chicago) for providing MOAB2 antibody, and P. Das for providing the g-secretaseinhibitor LY411575. We are grateful to D. Dickson, M. Castanedes Casey, L. Rousseau, and V. Phillipsfor histology and immunohistochemical analyses. We also thank M. Davis for the careful reading ofthis manuscript. Funding: This work was supported by NIH grants P01NS074969, R01AG027924,R01AG035355, R01AG046205, and P50AG016574, and grants from the Alzheimer’s Association andCure Alzheimer’s Fund (to G.B.); NIH grant R01NS088496 (to Y.Y.); NIH grant R01AG042513 (toJ.R.C.); Mayo Clinic CRM (Center for Regenerative Medicine) Career Development Award (to T.K.);and a fellowship from the American Heart Association (to C.-C.L.). Author contributions: C.-C.L.,T.K., and G.B. designed and set up the research concept. C.-C.L., N.Z., T.K., and G.B. designed andperformed the experiments. J.R.C. guided C.-C.L. to perform in vivo Abmicrodialysis experiments.D.M.H., Y.Y., and J.R.C. provided valuable technical and conceptual advice and experimentaltools. C.-C.L., T.K., and G.B. wrote the manuscript with critical inputs and edits by co-authors.Competing interests: D.H. is on the scientific advisory boards of Genentech, AstraZeneca,Neurophage, and Denali, and has consulted for Eli Lilly, AbbVie, Novartis, and Ono Pharma. D.H.is a cofounder of C2N Diagnostics LLC and has equity in the company. The other authors declarethat they have no competing interests.

Submitted 3 September 2015Accepted 11 March 2016Published 30 March 201610.1126/scitranslmed.aad3650

Citation: C.-C. Liu, N. Zhao, Y. Yamaguchi, J. R. Cirrito, T. Kanekiyo, D. M. Holtzman, G. Bu,Neuronal heparan sulfates promote amyloid pathology by modulating brain amyloid-bclearance and aggregation in Alzheimer’s disease. Sci. Transl. Med. 8, 332ra44 (2016).

eTranslationalMedicine.org 30 March 2016 Vol 8 Issue 332 332ra44 11

clearance and aggregation in Alzheimer's diseaseβNeuronal heparan sulfates promote amyloid pathology by modulating brain amyloid-

Chia-Chen Liu, Na Zhao, Yu Yamaguchi, John R. Cirrito, Takahisa Kanekiyo, David M. Holtzman and Guojun Bu

DOI: 10.1126/scitranslmed.aad3650, 332ra44332ra44.8Sci Transl Med

prevention and treatment.-HSPG interactions might be an effective strategy for ADβtissue. These findings suggest that targeting A

aggregation. Also, the authors found that several HSPG species were increased in human AD postmortem brain β and reduction in Aβforebrain neurons were protected from amyloid deposition because of a faster clearance of A

amyloid plaques. Here, Liu and colleagues show that genetically engineered mice lacking heparan sulfates indisease (AD). Heparan sulfate proteoglycans (HSPGs) are abundant cell surface receptors that colocalize with

) in the brain is a pathological hallmark of Alzheimer'sβ (AβThe accumulation of neurotoxic amyloid-Neuronal HSPGs lay a trap for amyloid

ARTICLE TOOLS http://stm.sciencemag.org/content/8/332/332ra44

MATERIALSSUPPLEMENTARY http://stm.sciencemag.org/content/suppl/2016/03/28/8.332.332ra44.DC1

CONTENTRELATED

http://stm.sciencemag.org/content/scitransmed/13/585/eabe6178.fullhttp://stm.sciencemag.org/content/scitransmed/12/571/eaba6334.fullhttp://stm.sciencemag.org/content/scitransmed/12/570/eaba1871.fullhttp://science.sciencemag.org/content/sci/354/6314/904.full

REFERENCES

http://stm.sciencemag.org/content/8/332/332ra44#BIBLThis article cites 80 articles, 28 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

registered trademark of AAAS. is aScience Translational MedicineScience, 1200 New York Avenue NW, Washington, DC 20005. The title

(ISSN 1946-6242) is published by the American Association for the Advancement ofScience Translational Medicine

Copyright © 2016, American Association for the Advancement of Science

by guest on March 26, 2021

http://stm.sciencem

ag.org/D

ownloaded from