Posts Tagged ‘A beta’

New Studies toward Understanding Alzheimer Disease

Curators: Larry H. Bernstein, MD, FCAP and Aviva Lev-Ari, PhD, RN


There is no unifying concept of Alzheimer Disease beyond the Tau and beta amyloid roles.  Recently, Ingenbleek and Bernstein (journal AD) made the connection between the age related decline of liver synthesis of plasma transthyretin and the more dramatic decline of transthyretin at the blood brain barrier, and the relationship to inability to transfer vitamin A via retinol binding protein to the brain.  Related metabolic events are reported by several groups.


What else is New?


Amyloid-β peptide protects against microbial infection in mouse and worm models of Alzheimer’s disease.

Kumar DK, Choi SH, Washicosky KJ, Eimer WA, Tucker S, Ghofrani J, Lefkowitz A, McColl G, Goldstein LE, Tanzi RE, Moir RD.

Sci Transl Med. 2016 May 25;8(340):340ra72.

They show that Aβ oligomerization, a behavior traditionally viewed as intrinsically pathological, may be necessary for the antimicrobial activities of the peptide. Collectively, our data are consistent with a model in which soluble Aβ oligomers first bind to microbial cell wall carbohydrates via a heparin-binding domain. Developing protofibrils inhibited pathogen adhesion to host cells. Propagating β-amyloid fibrils mediate agglutination and eventual entrapment of unatttached microbes….Salmonella Typhimurium bacterial infection of the brains of transgenic 5XFAD mice resulted in rapid seeding and accelerated β-amyloid deposition, which closely colocalized with the invading bacteria.

This is quite interesting in that infection drives the production of acute phase reactants resulting in decreased production of transthyretin.  Whether this also has ties to chronic disease in the elderly and risk of AD is not known.

Gain-of-function mutations in protein kinase Cα (PKCα) may promote synaptic defects in Alzheimer’s disease.

Alfonso SI, Callender JA, Hooli B, Antal CE, Mullin K, Sherman MA, Lesné SE, Leitges M, Newton AC, Tanzi RE, Malinow R.

Sci Signal. 2016 May 10;9(427):ra47.

Through whole-genome sequencing of 1345 individuals from 410 families with late-onset AD (LOAD), they identified three highly penetrant variants in PRKCA, the gene that encodes protein kinase Cα (PKCα), in five of the families. All three variants linked with LOAD displayed increased catalytic activity relative to wild-type PKCα as assessed in live-cell imaging experiments using a genetically encoded PKC activity reporter. Deleting PRKCA in mice or adding PKC antagonists to mouse hippocampal slices infected with a virus expressing the Aβ precursor CT100 revealed that PKCα was required for the reduced synaptic activity caused by Aβ. In PRKCA(-/-) neurons expressing CT100, introduction of PKCα, but not PKCα lacking a PDZ interaction moiety, rescued synaptic depression, suggesting that a scaffolding interaction bringing PKCα to the synapse is required for its mediation of the effects of Aβ. Thus, enhanced PKCα activity may contribute to AD, possibly by mediating the actions of Aβ on synapses.


Science Signaling Podcast for 10 May 2016: PKCα in Alzheimer’s disease.

Newton AC, Tanzi RE, VanHook AM.

Sci Signal. 2016 May 10;9(427):pc11. doi: 10.1126/scisignal.aaf9436.

Relevance of the COPI complex for Alzheimer’s disease progression in vivo.

Bettayeb K, Hooli BV, Parrado AR, Randolph L, Varotsis D, Aryal S, Gresack J,Tanzi RE, Greengard P, Flajolet M.

Proc Natl Acad Sci U S A. 2016 May 10;113(19):5418-23.

Inhibition of death-associated protein kinase 1 attenuates the phosphorylation and amyloidogenic processing of amyloid precursor protein.

Kim BM, You MH, Chen CH, Suh J, Tanzi RE, Ho Lee T.

Hum Mol Genet. 2016 Apr 19. pii: ddw114.

Extracellular deposition of amyloid-beta (Aβ) peptide, a metabolite of sequential cleavage of amyloid precursor protein (APP), is a critical step in the pathogenesis of Alzheimer’s disease (AD). While death-associated protein kinase 1 (DAPK1) is highly expressed in AD brains and its genetic variants are linked to AD risk, little is known about the impact of DAPK1 on APP metabolism and Aβ generation. This study demonstrated a novel effect of DAPK1 in the regulation of APP processing using cell culture and mouse models. DAPK1, but not its kinase deficient mutant (K42A), significantly increased human Aβ secretion in neuronal cell culture models. Moreover, knockdown of DAPK1 expression or inhibition of DAPK1 catalytic activity significantly decreased Aβ secretion. Furthermore, DAPK1, but not K42A, triggered Thr668 phosphorylation of APP, which may initiate and facilitate amyloidogenic APP processing leading to the generation of Aβ. In Tg2576 APPswe-overexpressing mice, knockout of DAPK1 shifted APP processing toward non-amyloidogenic pathway and decreased Aβ generation. Finally, in AD brains, elevated DAPK1 levels showed co-relation with the increase of APP phosphorylation. Combined together, these results suggest that DAPK1 promotes the phosphorylation and amyloidogenic processing of APP, and that may serve a potential therapeutic target for AD.

Recapitulating amyloid β and tau pathology in human neural cell culture models: clinical implications.

Choi SH, Kim YH, D’Avanzo C, Aronson J, Tanzi RE, Kim DY.

US Neurol. 2015 Fall;11(2):102-105.    Free PMC Article

The “amyloid β hypothesis” of Alzheimer’s disease (AD) has been the reigning hypothesis explaining pathogenic mechanisms of AD over the last two decades. However, this hypothesis has not been fully validated in animal models, and several major unresolved issues remain. Our 3D human neural cell culture model system provides a premise for a new generation of cellular AD models that can serve as a novel platform for studying pathogenic mechanisms and for high-throughput drug screening in a human brain-like environment.

The two key pathological hallmarks of AD are senile plaques (amyloid plaques) and neurofibrillary tangles (NFTs), which develop in brain regions responsible for memory and cognitive functions (i.e. cerebral cortex and limbic system) 3. Senile plaques are extracellular deposits of amyloid-β (Aβ) peptides, while NFTs are intracellular, filamentous aggregates of hyperphosphorylated tau protein 4.

The identification of Aβ as the main component of senile plaques by Drs. Glenner and Wong in 1984 5 resulted in the original formation of the “amyloid hypothesis.” According to this hypothesis, which was later renamed the “amyloid-β cascade hypothesis” by Drs. Hardy and Higgins 6, the accumulation of Aβ is the initial pathological trigger in the disease, subsequently leading to hyperphosphorylation of tau, causing NFTs, and ultimately, neuronal death and dementia 4,710. Although the details have been modified to reflect new findings, the core elements of this hypothesis remain unchanged: excess accumulation of the pathogenic forms of Aβ, by altered Aβ production and/or clearance, triggers the vicious pathogenic cascades that eventually lead to NFTs and neuronal death.

Over the last two decades, the Aβ hypothesis of AD has reigned, providing the foundation for numerous basic studies and clinical trials 4,7,10,11. According to this hypothesis, the accumulation of Aβ, either by altered Aβ production and/or clearance, is the initial pathological trigger in the disease. The excess accumulation of Aβ then elicits a pathogenic cascade including synaptic deficits, altered neuronal activity, inflammation, oxidative stress, neuronal injury, hyperphosphorylation of tau causing NFTs and ultimately, neuronal death and dementia 4,710.

One of the major unresolved issues of the Aβ hypothesis is to show a direct causal link between Aβ and NFTs 1214. Studies have demonstrated that treatments with various forms of soluble Aβ oligomers induced synaptic deficits and neuronal injury, as well as hyperphosphorylation of tau proteins, in mouse and rat neurons, which could lead to NFTs and neurodegeneration in vivo 1821. However, transgenic AD mouse models carrying single or multiple human familial AD (FAD) mutations in amyloid precursor protein (APP) and/or presenilin 1 (PS1) do not develop NFTs or robust neurodegeneration as observed in human patients, despite robust Aβ deposition 13,22,23. Double and triple transgenic mouse models, harboring both FAD and tau mutations linked with frontotemporal dementia (FTD), are the only rodent models to date displaying both amyloid plaques and NFTs. However, the NFT pathology in these models stems mainly from the overexpression of human tau as a result of the FTD, rather than the FAD mutations24,25.

Human neurons carrying FAD mutations are an optimal model to test whether elevated levels of pathogenic Aβ trigger pathogenic cascades including NFTs, since those cells truly share the same genetic background that induces FAD in humans. Indeed, Israel et al., observed elevated tau phosphorylation in neurons with an APP duplication FAD mutation 33. Blocking Aβ generation by β-secretase inhibitors significantly decreased tau phosphorylation in the same model, but γ-secretase inhibitor, another Aβ blocker, did not affect tau phosphorylation 33. Neurons with the APP V717I FAD mutation also showed an increase in levels of phospho tau and total tau levels 28. More importantly, Muratore and colleagues showed that treatments with Aβ-neutralizing antibodies in those cells significantly reduced the elevated total and phospho tau levels at the early stages of differentiation, suggesting that blocking pathogenic Aβ can reverse the abnormal tau accumulation in APP V717I neurons 28.

Recently, Moore et al. also reported that neurons harboring the APP V717I or the APP duplication FAD mutation showed increases in both total and phospho tau levels 27. Interestingly, altered tau levels were not detected in human neurons carrying PS1 FAD mutations, which significantly increased pathogenic Aβ42 species in the same cells 27. These data suggest that elevated tau levels in these models were not due to extracellular Aβ accumulation but may possibly represent a very early stage of tauopathy. It may also be due to developmental alterations induced by the APP FAD mutations.

As summarized, most human FAD neurons showed significant increases in pathogenic Aβ species, while only APP FAD neurons showed altered tau metabolism that may represent very early stages of tauopathy. However, all of these human FAD neurons failed to recapitulate robust extracellular amyloid plaques, NFTs, or any signs of neuronal death, as predicted in the amyloid hypothesis.

In our recent study, we moved one step closer to proving the amyloid hypothesis. By generating human neural stem cell lines carrying multiple mutations in APP together with PS1, we achieved high levels of pathogenic Aβ42 comparable to those in brains of AD patients 4446.

An external file that holds a picture, illustration, etc. Object name is nihms740236f1.jpg


Platform for AD drug screening in human neural progenitor cells with FAD mutations in a 3D culture system, which successfully reproduce human AD pathogenesis (amyloid plaques-driven tauopathy).

In addition to the impact on toxic Aβ species, our 3D culture model can test if these antibodies can block tau pathologies in 3D human neural cell culture systems 4446. Human cellular AD models can also be used to determine optimal doses of candidate AD drugs to block Aβ and/or tau pathology without affecting neuronal survival (Fig. 1).

While much progress has been made, many challenges still lie on the path to creating human neural cell culture models that comprehensively recapitulate pathogenic cascades of AD. A major difficulty lies in reconstituting the brain regions most affected in AD: the hippocampus and specific cortical layers. Recent progress in 3D culture technology, such as “cerebral organoids,” may also be helpful in rebuilding the brain structures that are affected by AD in a dish 52,53. These “cerebral organoids” were able to model various discrete brain regions including human cortical areas 52, which enabled them to reproduce microcephaly, a brain developmental disorder. Similarly, pathogenic cascades of AD may be recapitulated in cortex-like structures using this model. Adding neuroinflammatory components, such as microglial cells, which are critical in AD pathogenesis, will illuminate the validity of the amyloid β hypothesis. Reconstitution of robust neuronal death stemming from Aβ and tau pathologies will be the next major step in comprehensively recapitulating AD in a cellular model.


Family-based association analyses of imputed genotypes reveal genome-wide significant association of Alzheimer’s disease with OSBPL6, PTPRG, and PDCL3.

Herold C, Hooli BV, Mullin K, Liu T, Roehr JT, Mattheisen M, Parrado AR, Bertram L, Lange C, Tanzi RE.

Mol Psychiatry. 2016 Feb 2.

Relationship between ubiquilin-1 and BACE1 in human Alzheimer’s disease and APdE9 transgenic mouse brain and cell-based models.

Natunen T, Takalo M, Kemppainen S, Leskelä S, Marttinen M, Kurkinen KM, Pursiheimo JP, Sarajärvi T, Viswanathan J, Gabbouj S, Solje E, Tahvanainen E, Pirttimäki T, Kurki M, Paananen J, Rauramaa T, Miettinen P, Mäkinen P, Leinonen V, Soininen H, Airenne K, Tanzi RE, Tanila H, Haapasalo A, Hiltunen M.

Neurobiol Dis. 2016 Jan;85:187-205.

Accumulation of β-amyloid (Aβ) and phosphorylated tau in the brain are central events underlying Alzheimer’s disease (AD) pathogenesis. Aβ is generated from amyloid precursor protein (APP) by β-site APP-cleaving enzyme 1 (BACE1) and γ-secretase-mediated cleavages. Ubiquilin-1, a ubiquitin-like protein, genetically associates with AD and affects APP trafficking, processing and degradation. Here, we have investigated ubiquilin-1 expression in human brain in relation to AD-related neurofibrillary pathology and the effects of ubiquilin-1 overexpression on BACE1, tau, neuroinflammation, and neuronal viability in vitro in co-cultures of mouse embryonic primary cortical neurons and microglial cells under acute neuroinflammation as well as neuronal cell lines, and in vivo in the brain of APdE9 transgenic mice at the early phase of the development of Aβ pathology. Ubiquilin-1 expression was decreased in human temporal cortex in relation to the early stages of AD-related neurofibrillary pathology (Braak stages 0-II vs. III-IV). There was a trend towards a positive correlation between ubiquilin-1 and BACE1 protein levels. Consistent with this, ubiquilin-1 overexpression in the neuron-microglia co-cultures with or without the induction of neuroinflammation resulted in a significant increase in endogenously expressed BACE1 levels. Sustained ubiquilin-1 overexpression in the brain of APdE9 mice resulted in a moderate, but insignificant increase in endogenous BACE1 levels and activity, coinciding with increased levels of soluble Aβ40 and Aβ42. BACE1 levels were also significantly increased in neuronal cells co-overexpressing ubiquilin-1 and BACE1. Ubiquilin-1 overexpression led to the stabilization of BACE1 protein levels, potentially through a mechanism involving decreased degradation in the lysosomal compartment. Ubiquilin-1 overexpression did not significantly affect the neuroinflammation response, but decreased neuronal viability in the neuron-microglia co-cultures under neuroinflammation. Taken together, these results suggest that ubiquilin-1 may mechanistically participate in AD molecular pathogenesis by affecting BACE1 and thereby APP processing and Aβ accumulation.

Correction to Cathepsin L Mediates the Degradation of Novel APP C-Terminal Fragments.

Wang H, Sang N, Zhang C, Raghupathi R, Tanzi RE, Saunders A.

Biochemistry. 2015 Sep 22;54(37):5781. Epub 2015 Sep 8. No abstract available.

Massachusetts Alzheimer’s Disease Research Center: progress and challenges.

Hyman BT, Growdon JH, Albers MW, Buckner RL, Chhatwal J, Gomez-Isla MT, Haass C, Hudry E, Jack CR Jr, Johnson KA, Khachaturian ZS, Kim DY, Martin JB, Nitsch RM, Rosen BR, Selkoe DJ, Sperling RA, St George-Hyslop P, Tanzi RE, Yap L, Young AB, Phelps CH, McCaffrey PG.

Alzheimers Dement. 2015 Oct;11(10):1241-5. Epub 2015 Aug 19. No abstract available.

Alzheimer’s in 3D culture: challenges and perspectives.

D’Avanzo C, Aronson J, Kim YH, Choi SH, Tanzi RE, Kim DY.

Bioessays. 2015 Oct;37(10):1139-48. doi: 10.1002/bies.201500063. Epub 2015 Aug 7. Review.

Synaptotagmins interact with APP and promote Aβ generation.

Gautam V, D’Avanzo C, Berezovska O, Tanzi RE, Kovacs DM.

Mol Neurodegener. 2015 Jul 23;10:31. doi: 10.1186/s13024-015-0028-5.

Near-infrared fluorescence molecular imaging of amyloid beta species and monitoring therapy in animal models of Alzheimer’s disease.

Zhang X, Tian Y, Zhang C, Tian X, Ross AW, Moir RD, Sun H, Tanzi RE, Moore A, Ran C.

Proc Natl Acad Sci U S A. 2015 Aug 4;112(31):9734-9. doi: 10.1073/pnas.1505420112. Epub 2015 Jul 21.

A 3D human neural cell culture system for modeling Alzheimer’s disease.

Kim YH, Choi SH, D’Avanzo C, Hebisch M, Sliwinski C, Bylykbashi E, Washicosky KJ, Klee JB, Brüstle O, Tanzi RE, Kim DY.

Nat Protoc. 2015 Jul;10(7):985-1006. doi: 10.1038/nprot.2015.065. Epub 2015 Jun 11.

Cathepsin L Mediates the Degradation of Novel APP C-Terminal Fragments.

Wang H, Sang N, Zhang C, Raghupathi R, Tanzi RE, Saunders A.

Biochemistry. 2015 May 12;54(18):2806-16. doi: 10.1021/acs.biochem.5b00329. Epub 2015 Apr 28. Erratum in: Biochemistry. 2015 Sep 22;54(37):5781.

γ-Secretase modulators reduce endogenous amyloid β42 levels in human neural progenitor cells without altering neuronal differentiation.

D’Avanzo C, Sliwinski C, Wagner SL, Tanzi RE, Kim DY, Kovacs DM.

FASEB J. 2015 Aug;29(8):3335-41. doi: 10.1096/fj.15-271015. Epub 2015 Apr 22.

PLD3 gene variants and Alzheimer’s disease.

Hooli BV, Lill CM, Mullin K, Qiao D, Lange C, Bertram L, Tanzi RE.

Nature. 2015 Apr 2;520(7545):E7-8. doi: 10.1038/nature14040. No abstract available.


Read Full Post »

Alzheimer’s disease, snake venome, amyloid and transthyretin

Larry H. Bernstein, MD, FCAP, Curator



Significant points:

  • Alzheimer’s Disease is characterized by amyloid plaques
  • The plaques have amyloid beta and tau
  • Toxic proteins accumulate in AD
  • snake venome activates enzymes (Endothelin Converting Enzyme-1 and Neprilysin) that break down the plaques that are sufficient in non-AD brain
  • Aβ peptides derive from proteolytic processing of a large (695/770 amino acids) type 1 transmembrane glycoprotein known as amyloid beta precursor protein (APP)
  • a natural variant of Amyloid-β (Aβ) carrying the A2V substitution protects heterozygous carriers from AD by its ability to interact with wild-type Aβ, hindering conformational changes and assembly
  • aggregated Aβ species, particularly oligomeric assemblies, trigger a cascade of events that lead to hyperphosphorylation, misfolding and assembly of the tau protein with formation of neurofibrillary tangles
  • [Aβ1-6A2VTAT(D)] revealed strong anti-amyloidogenic effects in vitro and protected human neuroblastoma cells from Aβ toxicity
  • while both Aβ1-6A2V and Aβ1-6WT display a predominant coil configuration, Aβ1-6A2V shows a slightly higher propensity to form secondary structure motifs involving two to three residues
  • Aβ1-6A2VTAT(D) maintains the in vitro anti-amyloidogenic properties of Aβ1-6A2V(D)
  • Transthyretin (TTR) influences plasma Aβ by reducing its levels
  • Transthyretin (TTR) binds Aβ peptide, preventing its deposition and toxicity
  • TTR facilitated peptide internalization of Aβ1-42 uptake by primary hepatocytes
  • Brain permeability to TTR
  • TTR regulates LRP1 levels, suggesting that TTR uses this receptor to promote Aβ clearance


Snake venom may hold key to breaking down plaques that cause Alzheimer’s disease

March 2, 2016

Alzheimer’s disease, snake venome, amyloid and transthyretin


Snake venom may hold key to breaking down plaques that cause Alzheimer’s disease

A toxic protein called amyloid beta is thought to play a key role in the onset of Alzheimer’s disease. In healthy people, amyloid beta is degraded by enzymes as it forms. However, in patients with the disease, these enzymes appear unable to adequately perform their actions, causing the toxic protein to accumulate into plaque deposits, which many researchers consider leads to dementia.

One of the Holy Grails of the pharmaceutical industry has been to find a drug that stimulates these enzymes in people, particularly those who are in the early stages of dementia, when amyloid plaques are just starting to accumulate.

Monash researchers have discovered what could well be this elusive drug candidate– a molecule in snake venom that appears to activate the enzymes involved in breaking down the amyloid plaques in the brain that are the hallmark of Alzheimer’s disease. Dr Sanjaya Kuruppu and Professor Ian Smith from Monash University’s Biomedicine Discovery Institute have just published their research in Nature Scientific Reports.

Dr Kuruppu has spent most of his research life studying snake venoms, looking for drug candidates.  When he began researching Alzheimer’s disease he says that “snake venom was an obvious place for me to start.”

He was looking for a molecule that would stimulate the enzymes to break down the amyloid plaques.  What he found, when screening various snake venoms, was in fact one molecule with the ability to enhance the activity of two plaque degrading enzymes. This molecule was extracted from a venom of a pit viper found in South and Central America. Dr Kuruppu and his team have developed synthetic versions of this molecule. Initial tests done in the laboratory using human cells have shown it to have the same effects as the native version found in the snake venom.

Dr Kuruppu is one of the four researchers in Australia to receive funding from the National Foundation for Medical Research and Innovation to conduct further testing of this newly-identified molecule.

Explore further: Alzheimer protein’s structure may explain its toxicity

More information: A. Ian Smith et al. N-terminal domain of Bothrops asper Myotoxin II Enhances the Activity of Endothelin Converting Enzyme-1 and Neprilysin, Scientific Reports (2016).


N-terminal domain of Bothrops asper Myotoxin II Enhances the Activity of Endothelin Converting Enzyme-1 and Neprilysin

  1. Ian Smith, Niwanthi W. Rajapakse, Oded Kleifeld, Bruno Lomonte,…, Helena C. Parkington, James C. Whisstock & Sanjaya Kuruppu

Scientific Reports 6, Article number: 22413 (2016)


Neprilysin (NEP) and endothelin converting enzyme-1 (ECE-1) are two enzymes that degrade amyloid beta in the brain. Currently there are no molecules to stimulate the activity of these enzymes. Here we report, the discovery and characterisation of a peptide referred to as K49-P1-20, from the venom of Bothrops asper which directly enhances the activity of both ECE-1 and NEP. This is evidenced by a 2- and 5-fold increase in the Vmax of ECE-1 and NEP respectively. The K49-P1-20 concentration required to achieve 50% of maximal stimulation (AC50) of ECE-1 and NEP was 1.92 ± 0.07 and 1.33 ± 0.12 μM respectively. Using BLITZ biolayer interferometry we have shown that K49-P1-20 interacts directly with each enzyme. Intrinsic fluorescence of the enzymes change in the presence of K49-P1-20 suggesting a change in conformation. ECE-1 mediated reduction in the level of endogenous soluble amyloid beta 42 in cerebrospinal fluid is significantly higher in the presence of K49-P1-20 (31 ± 4% of initial) compared with enzyme alone (11 ± 5% of initial; N = 8, P = 0.005, unpaired t-test). K49-P1-20 could be an excellent research tool to study mechanism(s) of enzyme stimulation, and a potential novel drug lead in the fight against Alzheimer’s disease.

Metalloproteases play a central role in regulating many physiological processes and consequently abnormal activity of these enzymes contribute to a wide range of disease pathologies. These include cardiovascular1 and neurodegenerative disease2 as well as many types of cancers1. Inhibitors of metalloproteases are widely used in research applications with some also approved for use in the clinic. However, molecules which stimulate the activity of these enzymes are rarely encountered, and as such our understanding of the mechanism(s) behind enzyme stimulation remains poor. Stimulators of enzyme activity can provide novel insights into enzyme biology and potentially open up avenues for the design of a novel class of drugs. For instance, ECE-1 and NEP are two metalloproteases that degrade amyloid beta (Aβ), the accumulation of which is a hallmark of Alzheimer’s disease.

Therefore it is of great interest to regulate the production of, and more importantly, the degradation of Aβ by stimulating the activity of these enzymes2. This in turn could reverse, prevent or at least halt the progression of Alzheimer’s disease.

Previous studies using animal models of Alzheimer’s disease have shown that increasing the expression of ECE3 and NEP4 through DNA based techniques can have beneficial effects. However, DNA based approaches can pose challenges for clinical translation. Molecules which can directly stimulate the activity of ECE-1 and NEP, or increase their expression are more attractive alternatives. Several studies have reported on the presence of molecules which increase the expression of or activity of NEP5,6,7. However, there are no reports on molecules which stimulate the activity of ECE-1. For example, polyphenols in green tea have been reported to increase the activity of NEP in cell culture models5, while the neuroprotective hormone humanin has been shown to increase the expression of NEP in a mouse model of Alzheimer’s disease6. In addition, Kynurenic acid elevates NEP expression as well as activity in human neuroblastoma cultures and mouse cortical neurones7. Therefore this study aimed to identify a molecule which stimulates the activity of ECE-1. Here we report on the discovery of K49-P1-20, a 20 amino acid peptide from the venom of B. asper which stimulates the activity of both ECE-1 and NEP. The effect of this peptide on other closely related enzymes was also examined.

Identification of K49-P1-20

We screened venom from species across different geographical regions for their effects on ECE-1 activity. The venom from B. asper was found to stimulate the activity of ECE-1 (624 ± 27% of control; Fig. 1a). Fractionation of venom confirmed that ECE-1 stimulation was mediated by the previously isolated B. aspermyotoxin II (Fig. 1a), a lysine 49 (K49) type phospholipase A2 found in this venom which induces myonecrosis upon envenoming8. Digestion of B. asper myotoxin II with ArgC proteinase indicated that the stimulation of ECE-1 activity was mediated by its N-terminal region (Fig. 1a). The synthetic peptide K49-P1-34 corresponding to the N-terminal region mimicked the stimulator effects of B. asper myotoxin II (Fig. 1a,b). No significant difference in the activation was observed between peptides K49-P1-20 and K49-P1-34 (Fig. 1a). However, the level of stimulation observed in the presence of K49-P9-34 and inverted sequence of K49-P1-20 was significantly less compared with native K49-P1-20 (Fig. 1a). Further digestion of peptide K49-P1-20 resulted in a reduction in its ability to stimulate ECE-1 activity (Fig. 1c) indicating the importance of residues 1-20 for maximal stimulation of ECE-1 activity. Peptide K49-P1-20 failed to inhibit direct twitches of the chick biventer cervicis nerve muscle preparation, confirming its lack of myotoxic effects (Fig. 1d), in agreement with the previous mapping of toxicity determinants of B. asper myotoxin II to its C-terminal region9.

Figure 1

Figure 1


Discovery of K49-P1-20 (a) Comparison of ECE-1 stimulating effects of venom, B. asper myotoxin II, peptides K49-P1-20, K49-P1-34, K49-P9-34 and inverted K49-P1-20 (10 ng/μL); (b) Schematic showing the amino acid sequence of B. asper myotoxin II (ArgC mediated cleavage sites are indicated by arrows). The underlined sections correspond to the sequence of synthetic peptides tested for their effects on ECE-1 activity; (c) trypsin mediated cleavage of K49-P1-20 produces peptides K49-P1-7 and K49-P8-20 (cleavage sites indicated by arrows, top panel); the effect of K49-P1-20, peptides K49-P1-7 and K49-P8-20 on ECE-1 activity (bottom panel); (d) a representative trace showing the effect of K49-P1-20 (25 μg/mL) on direct twitches of the chick biventer cervices muscle. The arrow indicates the point of addition of peptide. *Significantly different than ECE-1 + peptide K49-P1-20, P < 0.05, unpaired t-test, n = 48.

Alanine scan

Alanine substitution of Leu(2) and Ile(9) failed to enhance ECE-1 activity, indicating their importance for stimulating ECE-1 (Fig. 3). Alanine substitution of Leu(2), Phe(3), Glu(4), Leu(10), Glu(12), Thr(13), Lys(15), Lys(19) and Ser(20) failed to enhance NEP activity, indicating their importance for stimulating NEP (Fig. 3).

Figure 3: Alanine scan.

A library of K49-P1-20 analogs were synthesised where each subsequent residue was replaced by an Ala. These analogs were tested for their ability to stimulate ECE-1 and NEP activity. The K49-P1-20 analogs are shown in the middle, with the Ala substitutions indicated in red. Closed bar denotes enzyme alone and the native peptide is indicated in blue *significantly different compared to enzyme alone; P < 0.05; One-way ANOVA; n = 4.

K49-P1-20 and enzyme interaction and conformational changes

BLITZ Biolayer interferometry

N-terminal biotinylation of K49-P1-20 had no significant effect on its ability to stimulate ECE-1 activity (Fig. 4a). Interaction of ECE-1 and NEP with biotinylated K49-P1-20 immobilised on a streptavidin biosensor was indicated by an increase in response units (nm) over time (Fig. 4b). The interaction was rapidly reversible. There was only a minimal interaction between each of the enzymes and biotinylated version of inverted K49-P1-20.


Figure 4: Association between K49-P1-20 and enzymes.

Figure 4

Figure 4

(a) Effect of N-terminal biotinylation of K49-P1-20 on the activity of ECE-1. (b) Representative traces obtained using Biolayer interferometry showing the level of interaction between enzymes and the biotinylated version of native or inverted K49-P1-20; representative traces showing the effect of K49-P1-20 on the intrinsic fluorescence of (c) ECE-1 and (d) NEP. Fluorescence of K49-P1-20 alone, and the sum of fluorescence intensities of K49-P1-20 and enzyme is also indicated.

K49-P1-20 stimulates ECE-1 activity in cerebrospinal fluid

K49-P1-20 (1–30 ng/μL) stimulated the activity of rhECE-1 in cerebrospinal fluid obtained from a patient with Alzheimer’s disease, as evidenced by the enhanced cleavage of bradykinin based QFS (Fig. 7a). Addition of stimulated ECE-1 to cerebrospinal fluid obtained from patients with Alzheimer’s disease (N = 8) resulted in a significant decrease (31 ± 4%) in the levels of endogenous soluble Aβ42 over 4 h, compared with the addition of non-stimulated ECE-1 (11 ± 5%; P = 0.005, unpaired t-testFig. 7b). This decrease was blocked by the ECE-1 specific inhibitor CGS35066 (Fig. 7b).

Figure 7: K49-P1-20 stimulates ECE-1 activity in cerebrospinal fluid

Figure 7

(a) the effect of K49-P1-20 (1–30 ng/μL) on the activity of rhECE-1(0.04–ng/μL) added to cerebrospinal fluid obtained from a patient with Alzheimer’s disease at post mortem. Enzyme activity was measured using the bradykinin based QFS. * & α significantly different compared to ECE-1 alone or K49-P1-20 (1 ng/μL) respectively; P < 0.001; n = 5; one-way ANOVA. (b) The effect of ECE-1 alone (0.04 ng/μL); ECE-1 incubated with K49-P1-20 (300 ng/μL); or ECE-1+ K49-P1-20 + ECE-1 inhibitor CGS35066 (500 nM), on the levels of endogenous Aβ42 in cerebrospinal fluid taken from a patient with Alzheimer’s disease at post-mortem was determined using a commercially available ELISA kit. Significantly different compared to *ECE-1 alone P = 0.005; or **ECE-1 + K49-P1-20, P = 0.009; unpaired t-test, N = 8–11.


ECE-1 and NEP are two closely related metalloproteases that play a key role in many physiological and pathophysiological processes2,15,16. A common substrate to both enzymes is Aβ which plays a key role in the pathogenesis of Alzheimer’s disease2,15,16,17,18. Previous studies have reported the discovery of molecules which increase NEP activity5,6,7. However, there are no reports on molecules that increase ECE-1 activity. Here we report on the discovery of a peptide named K49-P1-20 from the venom of B. asper which stimulates the activity of both ECE-1 and NEP. Interaction of K49-P1-20 with ECE-1 or NEP appears to induce a change in its conformation leading to an increase in activity. Unlike the molecules reported in previous studies which increase NEP expression and therefore cellular NEP activity5,6,7, K49-P1-20 appears to allosterically regulate the activity of ECE-1 and NEP.

Animal venoms have long been a source of lead compounds for future pharmaceuticals and research tools19,20. We therefore screened venoms of snakes found in different geographical regions to identify a molecule that modulates the activity of ECE-1, and found that the venom of B. asper stimulated ECE-1 activity. Initial fractionation of venom indicated that this effect was mediated by a toxin known as B. asper myotoxin II which induces myonecrosis following envenoming8. B. aspermyotoxin II belongs to a class of toxins known as Lysine 49 phospholipase A2 myotoxins21. Asp to Lys substitution at position 49 is a key structural feature of these toxins and their toxic effects are independent of the phospholipase A2 activity. Digestion of this toxin with ArgC proteinase indicated that stimulation of ECE-1 activity was mediated by its N-terminal domain. The use of synthetic peptides of varying length corresponding to this region confirmed that these effects were in fact mediated by its first 20 amino acids. Inverted sequence of K49-P1-20 failed to induce an increase in ECE-1 activity (136 ± 12 as % of ECE-1 alone; n = 3-4), indicating that the specific sequence of K49-P1-20 is critical for the observed effects. Further shortening of this peptide resulted in a loss of ECE-1 stimulating effects. K49-P1-20 therefore appears to possess the shortest optimum sequence required for ECE-1 stimulation and was used in all downstream studies. Previous studies have shown that myotoxic effects of B. asper myotoxin II are mediated by is C-terminal domain9. In agreement with this result, K49-P1-20 showed no myotoxicity in chick biventer cervicis muscle.

Compared with enzyme alone, K49-P1-20 also significantly enhanced the activity (expressed as % of control) of closely related enzyme NEP (1606 ± 29), and two other metalloproteases ACE-2 (145 ± 8) and IDE (292 ± 38). The level of ACE-2 and IDE stimulation was however significantly less compared with NEP, therefore indicating degree of specificity towards ECE-1 and NEP. All further studies therefore focused on the effect of K49-P1-20 on ECE-1 and NEP activity. K49-P1-20 increased the activity of ECE-1 and NEP in a concentration dependant manner. The increase in activity of both enzymes become evident at a K49-P1-20 concentration of 0.23 μM, or a peptide: enzyme molar ratio of 1:368. The high level of ECE-1 and NEP stimulation observed in response to K49-P1-20 is most likely the result of a common binding region for K49-P1-20 within these enzymes. ECE-1 and NEP in deed share 40% sequence homology22. However the potential sites of interaction between the enzymes and K49-P1-20 are best identified through structural biology approaches that take into account the secondary and tertiary structure of the enzymes.

Physical interaction between the activating molecule and enzyme is a common characteristic in the mechanisms of enzyme activation23. We used biolayer interferometry to probe possible physical interaction between K49-P1-20 and ECE-1 or NEP. N-terminal biotinylation of K49-P1-20 had no significant impact on its ability to stimulate ECE-1 activity, thus facilitating its use as a tool in research applications. Biotinylated K49-P1-20 immobilised on a streptavidin biosensor interacted directly with both ECE-1 and NEP as evidenced by the increase in response units over time. This interaction however was not observed with the biotinylated version of inverted K49-P1-20.

It is logical to assume that a conformational change that occurs following interaction with K49-P1-20 mediates the increase in enzyme activity. We investigated this by examining the effect of K49-P1-20 on the intrinsic fluorescence of ECE-1 and NEP. Fluorescence spectra of each enzyme in the presence of K49-P1-20 were distinct from that of enzyme alone. In addition, the sum of individual spectra for K49-P1-20 and ECE-1 or NEP failed to overlap with the spectra obtained by incubating K49-P1-20 with enzymes. This suggests that spectral changes that occur in the presence of K49-P1-20 is the likely result of a change in conformation of the enzymes, which in turn is a possible consequence of a direct interaction with K49-P1-20.


Tackling amyloidogenesis in Alzheimer’s disease with A2V variants of Amyloid-β

Giuseppe Di Fede, Marcella Catania, Emanuela Maderna, Michela Morbin,…,,Fabio Moda, Matteo Salvalaglio, Mario Salmona  & Fabrizio Tagliavini

Scientific Reports 6, Article number: 20949 (2016)


We developed a novel therapeutic strategy for Alzheimer’s disease (AD) exploiting the properties of a natural variant of Amyloid-β (Aβ) carrying the A2V substitution, which protects heterozygous carriers from AD by its ability to interact with wild-type Aβ, hindering conformational changes and assembly thereof. As prototypic compound we designed a six-mer mutated peptide (Aβ1-6A2V), linked to the HIV-related TAT protein, which is widely used for brain delivery and cell membrane penetration of drugs. The resulting molecule [Aβ1-6A2VTAT(D)] revealed strong anti-amyloidogenic effects in vitro and protected human neuroblastoma cells from Aβ toxicity. Preclinical studies in AD mouse models showed that short-term treatment with Aβ1-6A2VTAT(D) inhibits Aβ aggregation and cerebral amyloid deposition, but a long treatment schedule unexpectedly increases amyloid burden, although preventing cognitive deterioration. Our data support the view that the AβA2V-based strategy can be successfully used for the development of treatments for AD, as suggested by the natural protection against the disease in human A2V heterozygous carriers. The undesirable outcome of the prolonged treatment with Aβ1-6A2VTAT(D) was likely due to the TAT intrinsic attitude to increase Aβ production, avidly bind amyloid and boost its seeding activity, warning against the use of the TAT carrier in the design of AD therapeutics.

Alzheimer’s disease (AD) is the most common form of dementia in the elderly. Its clinical course is slow but irreversible since no disease-modifying treatments are currently available. As a result, this illness has a huge socio-sanitary impact and designing of effective therapies is considered a public health priority.

A central pathological feature of AD is the accumulation of misfolded Amyloid-beta (Aβ) peptides in the form of oligomers and amyloid fibrils in the brain1,2,3. It has been advanced that aggregated Aβ species, particularly oligomeric assemblies, trigger a cascade of events that lead to hyperphosphorylation, misfolding and assembly of the tau protein with formation of neurofibrillary tangles and disruption of the neuronal cytoskeleton, widespread synaptic loss and neurodegeneration. According to this view, altered Aβ species are the primary cause of AD and the primary target for therapeutic intervention3,4.

Aβ peptides derive from proteolytic processing of a large (695/770 amino acids) type 1 transmembrane glycoprotein known as amyloid beta precursor protein (APP). APP is cleaved at the N-terminus of the Aβ domain by β-secretase, forming a large, soluble ectodomain (sAPPβ) and a 99-residue, membrane-retained C-terminal fragment (C99). Subsequently, γ-secretase cleaves C99 to release Aβ with different carboxyl termini, including Aβ40, Aβ42 and other minor species5. APP may undergo an alternative, non-amyloidogenic processing where the protein is cleaved within the Aβ domain by α-secretase, forming a soluble ectodomain (sAPPα) and an 83-residue C-terminal fragment (C83)5,6.

We identified a novel mutation in the APP gene resulting in A-to-V substitution at codon 673, corresponding to position 2 in the Aβ sequence7. Studies on biological samples from an A673V homozygous carrier, and cellular and C. elegans models indicated that this mutation shifts APP processing towards the amyloidogenic pathway with increased production of amyloidogenic peptides. Furthermore, the A2V substitution in the Aβ sequence (AβA2V) increases the propensity of the full-length peptides (i.e., Aβ1-40 and Aβ1-42) to adopt a β-sheet structure, boosts the formation of oligomers both in vitroand in vivo and enhances their neurotoxicity8,9,10. Following the observation that humans carrying the mutation in the heterozygous state do not develop AD, we carried out in vitro studies with synthetic peptides that revealed the extraordinary ability of AβA2V to interact with wild-type Aβ (AβWT), interfering with its nucleation or nucleation-dependent polymerization7. This provides grounds for developing a disease-modifying therapy for AD based on modified AβA2V peptides retaining the key functional properties of parental full-length AβA2V.

Following this approach, we generated a mutated six-mer peptide (Aβ1-6A2V), constructed entirely by D-amino acids [Aβ1-6A2V(D)] to increase its stability in vivo, whose interaction with full-length AβWT hinders oligomer production and prevents amyloid fibril formation8.

These results prompted us to develop a prototypic compound by linking Aβ1-6A2V(D) to an all-D form of TAT sequence [TAT(D)], a peptide derived from HIV that powerfully increases virus transmission to neighbour cells11, and is widely used for brain delivery of drugs12,13,14. Here we report that this compound [Aβ1-6A2VTAT(D)] has strong anti-amyloidogenic effects in vitro, leading to inhibition of oligomer, amyloid fibril formation and of Aβ-dependent neurotoxicity. Preclinical studies showed that a short-term treatment with this peptide in an AD mouse model prevents Aβ aggregation and amyloid deposition in the brain but longer treatment unexpectedly increases amyloid burden, most likely due to the TAT intrinsic attitude to enhance Aβ production and to avidly bind amyloid and boost its seeding activity, warning against the use of this carrier in therapeutic approaches for AD.

In silico molecular modeling of AβA2V peptide variants

To predict the structural basis of the anti-amyloidogenic effect of Aβ1-6A2V(D), a comparative conformation analysis of WT and mutated Aβ1-6 was carried out with all-atom classical MD simulations in explicit solvent. Both Aβ1-6WT and Aβ1-6A2V are intrinsically disordered peptides characterized by high flexibility. Nevertheless, the substitution of Ala2 with a Val residue induces significant changes in the appearance of the peptide in solution, resulting in an increase of the apolar character of the solvent accessible surface (SAS) (Fig. 1A) and in a modification of the gyration radius distribution in the Aβ1-6A2V. Figure 1B shows that the probability distribution of the gyration radius is characterized by a global shift to smaller values and by the appearance of a shoulder in the distribution corresponding to gyration radii of 0.5 nm.

Figure 1: Analysis of 1.5 μs explicit solvent MD simulations of the Aβ1-6WT and Aβ1-6A2V peptides.

An external file that holds a picture, illustration, etc. Object name is srep20949-f1.jpg

(A) Apolar character of the peptide SAS represented as the ratio between SASapolar and the total SAS. (B) Gyration radius distribution. (C) Analysis of secondary structure propensity. “Structure” indicates residues possessing a defined secondary structure, in this case structure indicates residues in a “turn” configuration. “Coil” indicates residues that do not display a defined secondary structure. Analysis of the secondary structure was carried out with DSSP. (D) Typical compact “turn” and elongated “coil” configurations reported for the Aβ1-6A2V and Aβ1-6WT, respectively. (E) Analysis of the most populated structural clusters. Representative structures of the six most probable clusters were reported. The coil configuration has been highlighted in green, the turn in red and a partly folded turn in orange.

An analysis of the secondary structure content displayed by the peptides (Fig. 1C) shows that, while both Aβ1-6A2Vand Aβ1-6WT display a predominant coil configuration, Aβ1-6A2V shows a slightly higher propensity to form secondary structure motifs involving two to three residues. Aβ1-6A2V in fact displays a propensity to form a turn involving the Glu3, Phe4 and Arg5 residues (Fig. 1D). The most populated structural clusters15 (Fig. 1E), in Aβ1-6WT are characterized by an elongated coil structure accounting for 52.6% of the configurations, while the compact “turn” state is only the third most probable cluster, with a population of around 9%. Conversely, in the Aβ1-6A2V, while the most populated structure is still an elongated coil (32%), the “turn” configuration is the second most populated structural cluster (31%).

Both Aβ1-6WT and Aβ1-6A2V under physiological conditions are characterized by intramolecular salt bridges such as those between Asp1 and Arg5 or Glu3-Arg5. In the extended coil configuration (Fig. 1E), salt bridges can be dynamically formed and dissociated without requiring a specific rearrangement of the peptide backbone. However, in the turn configuration salt bridges are typically dissociated; the interaction of the apolar Val2 sidechain with the Arg5 sidechain stabilizes such a dissociated state. The additional sterical hindrance to the rearrangement induced by the Val2 sidechain also contributes to the stabilization of the turn configuration of the A2V peptide.

The propensity of the A2V mutant to adopt a Glu3-Arg5 turn configuration characterized by a significant lifetime can be interpreted as the probable source of the heterotypic interaction of the Aβ1-6A2V with full-length Aβ, which results in hindering its assembly.

Aβ1-6A2V retains the in vitro anti-amyloidogenic features of the parental full-length peptide


We previously showed that Aβ1-6A2V(D) destabilizes the secondary structure of Aβ1-42WT8 and is even more effective than the WT peptide [Aβ1-6WT(D)] and the A2V-mutated L-isomer [Aβ1-6A2V(L)] at preventing the aggregation of full-length AβWT8.

Treatment of SH-SY5Y cells with Aβ1-6WT(D) or Aβ1-6A2V(D) showed that neither is toxic for living cells even at high concentrations (20 μM) (Fig. 2A,B) and that both peptides are able to reduce the toxicity induced by Aβ1-42WT (Fig. 2C,D). However, Aβ1-6A2V(D) showed a stronger effect in counteracting the reduction of cell viability caused by Aβ1-42WT (Fig. 2D), suggesting that the A-to-V substitution actually amplifies the protective effects of the six-mer peptide.

Figure 2: Analysis of the effects of Aβ1-6WT(D), Aβ1-6A2V(D) and Aβ1-6A2VTAT(D) on neurotoxicity in cell models.

SH-SY5Y cells were differentiated with 10 μM retinoic acid. After 6 days the proper peptide was added to culture medium and cell viability was assessed after 24 h by MTT test. (A,B) Neither Aβ1-6WT(D) nor Aβ1-6A2V(D) are significantly toxic when added to culture medium of differentiated SH-SY5Y cells. Conversely, Aβ1-42WT reduces cell viability by 35%. * Significance vs non-treated cells. (C,D) Both Aβ1-6WT(D) and Aβ1-6A2V(D) are able to counteract the toxic effect of Aβ1-42WT. Aβ1-6A2V(D) showed a stronger effect than Aβ1-6WT(D). (E) Aβ1-6A2VTAT(D) is not toxic when added to culture medium at concentrations ranging between 1 and 5 μM, while it reduces cell viability at higher concentrations. * Significance vs non-treated cells. (F) Aβ1-6A2VTAT(D) showed a dose-dependent effect in reducing Aβ1-42wt toxicity. Comparison of cell viability was performed by Student t-test.

Aβ1-6A2VTAT(D) maintains the in vitro anti-amyloidogenic properties of Aβ1-6A2V(D)

Aβ1-6A2V(D) alone does not efficiently cross either the blood brain barrier (BBB) or cell membranes (data not shown). This is an important feature that would deeply limit its use as an in vivo anti-amyloidogenic drug. So, we linked this peptide to the all-D TAT sequence to improve the translocation of Aβ1-6A2V(D) across the BBB and cell membranes, minimize the degradation of the peptide and reduce the immune response elicited by the molecule. The resulting compound [Aβ1-6A2VTAT(D)] destabilizes the secondary structure of Aβ1-42WT. Indeed, CD spectroscopy studies showed that Aβ1-6A2VTAT(D) inhibits the acquisition of β-sheet conformation by Aβ1-42WT (data not shown), thus affecting the folding of the full-length peptide.

We tested the ability of Aβ1-6A2VTAT(D) to inhibit the fibrillogenic properties of the full-length Aβ in vitro and found that the compound hindered Aβ1-42WT aggregation (Fig. 3). Polarized light and electron microscopy studies on aggregates of Aβ1-42WT formed after 20 days incubation with or without Aβ1-6A2VTAT(D) revealed that the mutated peptide hinders the formation of amyloid structures (Fig. 3B) and reduces the amount of fibrils generated by the full-length peptide (Fig. 3D). Moreover, AFM analysis (Fig. 3E,H) showed that Aβ1-6A2VTAT(D) actually interferes with the oligomerization process of Aβ1-42WT. Indeed, monomeric Aβ1-42WT, incubated alone at a final concentration of 100 μM, formed a family of small oligomers of different size within a range of 6-20 nm in diameter (~ 70%) (Fig. 3E,G). Conversely, the co-incubation with Aβ1-6A2VTAT(D) resulted in the formation of very small globular structures with a range of 5-8 nm in diameter and height of 200-400 pm (~ 70%), large and thin structures, apparently very rich in water (width: 500–700 nm; height: 200–500 pm). Notably, only rare oligomeric structures were detected (Fig. 3F,H).

Figure 3: Inhibition of aggregation of Aβ1-42WT by Aβ1-6A2VTAT(D).

Figure 3

Polarized-light (A,B), electron microscopy (C,D) and atomic force microscopy (AFM) (E–H) studies showing the inhibitory effects of Aβ1-6A2VTAT(D) on amyloid formation, fibril production and oligomerization by Aβ1-42WT. In polarized-light and EM studies, both peptides were used at 0.125 mM, molar ratio = 1:1 or 1:4 respectively, with 20 days incubation. From 5–20 days, 1:1 co-incubation of the two peptides (B,D) displayed a lower amyloid fibril content respect to Aβ1-42WT alone (A,C), showing protofibrils, short fibrils and disaggregated granular material.E,F: Representative Tapping mode of AFM images as determined by amplitude error data of Aβ1-42WT oligomers. Aβ1-42WT peptide 100 μM in phosphate buffer 50 mM, pH 7.4 was incubated at 4 °C for 24 h alone (E) (Z range: -10/ + 10 mV) or in presence of Aβ1-6A2VTAT(D) (F) (Z range: -10/ + 25 mV). The molar ratio of Aβ1-42WT to Aβ1-6A2VTAT(D) was 1:4. Scale bar: 1 μm, inset: 200 nm. (G,H): height plot profiles obtained along different lines traced on the topographic AFM images. Overall, these effects were already evident in the 1:1 mixture of the two peptides (data not shown), suggesting that the inhibition of Aβ1-42WT aggregation by Aβ1-6A2VTAT(D) is a dose-dependent effect.

These effects were observed by incubating Aβ1-42WT and Aβ1-6A2VTAT(D) at a 1:4 molar ratio, but they were also evident at equimolar concentrations of the two peptides.

Moreover, treatment of differentiated SH-SY5Y cells with Aβ1-6A2VTAT(D) showed that the peptide is not toxic when administered at concentrations ranging between 1 and 5 μM (Fig. 2E). When co-incubated with Aβ1-42WT, Aβ1-6A2VTAT(D) displayed a significant dose-dependent reduction of the toxicity induced by full-length Aβ (Fig. 2F).

All these findings indicated that the designed Aβ1-6A2VTAT(D) peptide is particularly efficient at inhibiting Aβ polymerization and toxicity in vitro, and identified it as our lead compound for the subsequent in vivo studies.

During the last few decades, huge efforts have been made to develop disease-modifying therapies for Alzheimer, but the results of these attempts have been frustrating. The anticipated increase of AD patients in the next few decades makes the development of efficient treatments an urgent issue16. In order to prevent the disease and radically change its irreversible course, a long series of experimental strategies against the main molecular actors of the disease (Aβ and tau)17 or novel therapeutic targets18 have been designed based on purely theoretical grounds19 as well as on evidence mainly deriving from preclinical observations in AD animal models20. However, few strategies proved suitable for application in human clinical trials, and none proved to be really effective21.Our approach differs from previous strategies – mainly those involving modified Aβ peptides that have been found to inhibit amyloidogenesis19,22 – since it is based on a natural genetic variant of amyloid-β (AβA2V) that occurs in humans and prevents the development of the disease when present in the heterozygous state7.

In this context, we carried out in vitro and in vivo studies that revealed the extraordinary ability of AβA2V to interact with AβWT, interfering with its aggregation8. These findings were a proof of concept of the validity of therapeutic strategies based on the use of AβA2V variant, and prompted us to develop a new disease-modifying treatment for AD by designing a six-mer mutated D-isomer peptide [Aβ1-6A2V(D)] linked to the short amino acid sequence derived from the HIV TAT peptide, widely used for brain delivery, to make the translocation of Aβ1-6A2V(D) across the BBB feasible.

The use of TAT as a carrier for brain delivery of drugs has been employed in several experimental approaches for the treatment of AD-like pathology in mouse models12,13. Recently, intraperitoneal administration of a TAT-BDNF peptide complex for 1 month was shown to improve the cognitive functions in AD rodent models23.

A previous study showed that, following its peripheral injection, a fluorescein-labelled version of TAT is able to cross the BBB, bind amyloid plaques and activate microglia in the cerebral cortex of APPswe/PS1DE9 transgenic mice24. TAT was then conjugated with a peptide inhibitor (RI-OR2, Ac-rGffvlkGr-NH2) consisting of a retro-inverted version of Aβ16–20 sequence25 that was found to block the formation of Aβ aggregates in vitro and to inhibit the toxicity of Aβ on cultured cells25. Daily i.p. injection of RI-OR2-TAT for 21 days into 10-month-old APPswe/PS1DE9 mice resulted in a reduction in Aβ oligomer levels and amyloid-β burden in cerebral cortex24.

We followed a similar strategy and initially demonstrated that Aβ1-6A2V(D), with or without the TAT sequence, retains in vitro the anti-amyloidogenic properties of the parental full-length mutated Aβ, since it is effective at hindering in vitro the production of oligomers and fibrils, the formation of amyloid and the toxicity induced by Aβ1-42WT peptide on SYSH-5Y cells.

Based on these results, we then decided to test in vivothe anti-amyloidogenic ability of Aβ1-6A2VTAT(D). The compound proved stable in serum after i.p. administration in mice, able to cross the BBB and associated with an immune response that was not found to cause any brain damage.

Short-term treatment with Aβ1-6A2VTAT(D) in the APPswe/PS1DE9 mouse model prevented cognitive deterioration, Aβ aggregation and amyloid deposition in brain. Unexpectedly, a longer treatment schedule, while retaining the results for cognitive impairment, attenuated the effects on Aβ production and increased amyloid burden, most likely due to the intrinsic amyloidogenic properties of TAT.


Indeed, we found that TAT(D), unlike Aβ1-6A2V(D), has a strong ability to bind amyloid deposits. This avidity for amyloid could boost the intrinsic seeding activity of amyloid plaques via a continuous and self-sustained recruitment of Aβ aggregates, leading to an exacerbation of the amyloidogenesis.

A similar effect of TAT was described in a study26reporting that HIV TAT promotes AD-like pathology in an AD mouse model co-expressing human APP bearing the Swedish mutation and TAT peptide (PSAPP/TAT mice). These mice indeed showed more Aβ deposition, neurodegeneration, neuronal apoptotic signalling, and phospho-tau production than PSAPP mice.

Moreover, TAT was found to increase Aβ levels by inhibiting neprilysin27 or enhancing β-secretase cleavage of APP, resulting in increased levels of the C99 APP fragment and 5.5-fold higher levels of Aβ4228. The same study reported that stereotaxic injection of a lentiviral TAT expression construct into the hippocampus of APP/presenilin-1 (PS1) transgenic mice resulted in increased TAT-mediated production of Aβ in vivo as well as an increase in the number and size of Aβ plaques. This is consistent with our findings, indicating a shift in APP processing towards the amyloidogenic processing in vivo at the end of the 5-month treatment with Aβ1-6A2VTAT(D) that was not observed in shorter treatment schedules with the same compound.

Therefore, these data suggest that the final outcome of our in vivo studies with Aβ1-6A2VTAT(D) is the result of side effects of the TAT carrier, whose amyloidogenic intrinsic activity neutralized the anti-amyloidogenic properties of the AβA2V variant. Nevertheless, we believe that the approach based on the use of AβA2V variant can be successfully used in treating AD, because of its potential ability to tackle the main pathogenic events involved in the disease, as suggested by the natural protection against the disease which occurs in human heterozygous A673V carriers.


Transthyretin participates in beta-amyloid transport from the brain to the liver- involvement of the low-density lipoprotein receptor-related protein 1?

Mobina Alemi, Cristiana Gaiteiro, Carlos Alexandre Ribeiro, Luís Miguel Santos,João Rodrigues Gomes,…, Ignacio Romero, Maria João Saraiva  & Isabel Cardoso

Scientific Reports 6, Article number: 20164 (2016)

Transthyretin (TTR) binds Aβ peptide, preventing its deposition and toxicity. TTR is decreased in Alzheimer’s disease (AD) patients. Additionally, AD transgenic mice with only one copy of the TTR gene show increased brain and plasma Aβ levels when compared to AD mice with both copies of the gene, suggesting TTR involvement in brain Aβ efflux and/or peripheral clearance. Here we showed that TTR promotes Aβ internalization and efflux in a human cerebral microvascular endothelial cell line, hCMEC/D3. TTR also stimulated brain-to-blood but not blood-to-brain Aβ permeability in hCMEC/D3, suggesting that TTR interacts directly with Aβ at the blood-brain-barrier. We also observed that TTR crosses the monolayer of cells only in the brain-to-blood direction, as confirmed by in vivo studies, suggesting that TTR can transport Aβ from, but not into the brain. Furthermore, TTR increased Aβ internalization by SAHep cells and by primary hepatocytes from TTR+/+ mice when compared to TTR−/− animals. We propose that TTR-mediated Aβ clearance is through LRP1, as lower receptor expression was found in brains and livers of TTR−/− mice and in cells incubated without TTR. Our results suggest that TTR acts as a carrier of Aβ at the blood-brain-barrier and liver, using LRP1.

Alzheimer’s disease (AD), described for the first time by Alois Alzheimer in 1906, is characterized by progressive loss of cognitive functions ultimately leading to death1. Pathologically, the disease is characterized by the presence of extraneuronal amyloid plaques consisting of aggregates of amyloid-beta (Aβ) peptide, and neurofibrillary tangles (NFTs) which are intracellular aggregates of abnormally hyperphosphorylated tau protein2. Aβ peptide is generated upon sequential cleavage of the amyloid precursor protein (APP), by beta- and gamma-secretases, and it is believed that an imbalance between Aβ production and clearance results in its accumulation in the brain.

Clearance of Aβ from the brain occurs via active transport at the blood-brain-barrier (BBB) and blood cerebrospinal fluid (CSF) barrier (BCSFB), in addition to the peptidolytic removal of the peptide by several enzymes. The receptors for Aβ at the BBB bind Aβ directly, or bind to one of its carrier proteins, and transport it across the endothelial cell. The low-density lipoprotein receptor-related protein 1 (LRP1) and the receptor for advanced glycation end products (RAGE) are involved in receptor-mediated flux of Aβ across the BBB3. Both LRP1 and RAGE are multi-ligand cell surface receptors that, in addition to Aβ, mediate the clearance of a large number of proteins. While LRP1 appears to mediate the efflux of Aβ from the brain to the periphery, RAGE has been strongly implicated in Aβ influx back into the central nervous system (CNS). With increasing age, the expression of the Aβ efflux transporters is decreased and the Aβ influx transporter expression is increased at the BBB, adding to the amyloid burden in the brain.


Transthyretin (TTR), a 55 kDa homotetrameric protein involved in the transport of thyroid hormones and retinol, has been proposed as a protective protein in AD in the mid-nineties, when Schwarzman and colleagues described this protein as the major Aβ binding protein in CSF. These authors described that TTR was able to inhibit Aβ aggregation and toxicity, suggesting that when TTR fails to sequester Aβ, amyloid formation occurs4,5. Data showing that TTR is decreased in both CSF6 and plasma7,8 of AD patients, strengthen the idea of neuroprotection by TTR. Evidence coming from in vivostudies in AD transgenic mice established in different TTR genetic backgrounds9,10 also suggests that TTR prevents Aβ deposition and protects against neurodegeneration, although the exact mechanism is still unknown. Ribeiro and colleagues reported increased Aβ levels in both brain and plasma of AD mice with only one copy of the TTR gene, when compared to animals with two copies of the gene11, suggesting a role for TTR in Aβ clearance. Growing evidence also suggests a wider role for TTR in CNS neuroprotection, including in ischemia12, regeneration13 and memory14.

The presence of TTR in brain areas other than its site of synthesis and secretion – the choroid plexus (CP) and CSF, respectively–in situations of injury, such as ischemia, has been shown using a mouse model with compromised heat-shock response12. Authors showed that TTR was not being locally synthesized, but instead should derive from CSF TTR. However, other studies demonstrated TTR synthesis by cortical15 or hippocampal neurons both in vitro16, and in vivo17, and some hints on its regulation have already been advanced. Kerridge and colleagues showed that TTR is expressed in SH-SY5Y neuroblastoma cell line, and that it is up-regulated by the AICD fragment of amyloid precursor protein (APP), specifically derived from the APP695 isoform. Induced accumulation of functional AICD resulted in TTR up-regulation and Aβ decreased levels16. Wang and colleagues reported that TTR expression in SH-SY5Y cells, primary hippocampal neurons and hippocampus of APP23 mice is significantly enhanced by heat shock factor 1 (HSF1)17. In any case, TTR is available in the brain and might participate in brain Aβ efflux by promoting BBB permeability to the peptide. With regard to Aβ peripheral elimination, it is known that Aβ bound to ApoE/cholesterol can be incorporated in HDL to be further delivered at the liver for degradation18 and curiously, a fraction of TTR is transported in HDL19. Furthermore, the liver is the major site for TTR degradation and although its hepatic receptor has never been unequivocally identified, it has been reported that it is a RAP-sensitive receptor20. Thus, in this work we assessed the role of TTR in Aβ transport, both from the brain and to the liver.

TTR clearance in vivo

TTR ability to cross the BBB, in both directions, was studied in vivo using TTR −/− mice and injecting h rTTR. To assess the brain-to-blood permeability, immediately before the injection, mice were weighed and anesthetized with intraperitoneal injection of an anesthetic combination of ketamine and medetomidine (7.5 mg/Kg and 0.1 mg/Kg, respectively) and placed in a stereotaxic apparatus (Stoelting Co.). The cranium was exposed using an incision in the skin and one small hole was drilled through the cranium over the right lateral ventricle injection site to the following coordinates: mediolateral −1.0 mm, anterior-posterior −0.22 mm and dorsal-ventral −1.88 mm, from bregma. Then, 10 μg of h rTTR were injected into the brain using a 10 μL motorized syringe (Hamilton Co.) connected to a 30 gauge needle (RN Needle 6 pK, Hamilton Co.) at a rate of 0.75 μL/min (4 μL final volume). After injection, the microsyringe was left in place for 3 minutes to minimize any backflow, and then the incision was closed with sutures (Surgicryl), and the wound was cleaned with 70% ethanol. After surgery, the animals were kept warm, using a warming pad, and blood samples were collected by the tail vein after 20, 40 and 60 minutes, in a capillary tube (previously coated with EDTA). At the time of sacrifice (after 60 minutes), the mice were re-anesthetized with 75 mg/Kg ketamine and 1 mg/Kg medetomidine, and after total absence of reflexes in the paw and tail, mice were perfused through the injection of sterile PBS pH 7.4 via the inferior vena cava until the liver becomes blanched. Then, the brain was rapidly collected and frozen at −80 °C until use.

To assess the blood-to-brain permeability, 10 μg of h rTTR were injected in the tail vein, and blood samples were collected after 20, 40 and 60 minutes. At 60 minutes, and after perfusion as described above, CSF and brain were also collected.

To determine TTR levels, brains were weighted and homogenized in 750 μL of 50 mM TBS pH 7.4 containing protease inhibitor cocktail. After centrifugation for 20 minutes at 14000 rpm at 4 °C, supernatants were collected. TTR concentration in brain, CSF and plasmas was determined by ELISA.

Characterization of the hCMEC/D3 cell line

The hCMEC/D3 cell line represents a valid and powerfulin vitro tool as a BBB model, and presents a less expensive and more logistically feasible alternative to primary hBMEC cells24,25. Thus, our first step was the validation of the hCMEC/D3 model by characterizing this cell line regarding two critical features for our studies: BBB integrity and LRP1 expression.

In the context of endothelial cell tight junctions (TJ), hCMEC/D3 cells were tested for claudin-5 and occludin expression by immunofluorescence. As shown in Fig. 1, hCMEC/D3 cells are positive for TJ structural proteins, claudin-5 and occludin, showing the expected membrane localization (as previously described). These results indicate that the integrity, tightness and structure, as well as the paracellular contact between endothelial cells are guaranteed by these TJ proteins. Along with other TJ proteins expressed by hCMEC/D3, claudin-5 and occludin ensure, with high efficiency, the control of transport across the cells monolayer.

Figure 1: Immunofluorescence localization of TJs components Claudin-5 and Occludin, and of LRP1, in hCMEC/D3.


Figure 1

The expression of the efflux transport receptor LRP1 by the hCMEC/D3 cell line is a key factor when validating this model, both for BBB studies purposes and for Aβ transport research. Thus, we performed immunofluorescence analysis to verify if LRP1 exists in the hCMEC/D3 cells. Our results show that LRP1 is expressed in these cells ensuring the Aβ transport through the cells monolayer (Fig. 1).

Effect of TTR in Aβ1-42 internalization by hCMEC/D3

Aβ1-42 is transported across the BBB, as expected, and is internalized by hCMEC/D3 cells. We firstly investigated FAM-labelled Aβ1-42 (FAM-Aβ1-42, 500 ng/mL)) uptake by these cells in the absence and presence of human recombinant TTR (h rTTR) (7.5 μg/mL), and analysed the results by flow cytometry.

Cells were incubated with FAM-Aβ1-42 at 37 °C producing a rapid uptake of the peptide (Fig. 2A). After 5 minutes of incubation, 35–39% of the cells were fluorescent and after an additional 5 minutes (10 minutes incubation) a significant increase was already measured as over 57% of the cells were fluorescent, although differences between the presence and absence of TTR were not significant. However, after 15 minutes the presence of TTR significantly increased Aβ internalization resulting in about 73% fluorescent cells, in contrast to 61.7% incubated in the absence of TTR (Fig. 2A). Finally after 30 minutes of incubation, and although the difference between internalization levels at 15 and 30 minutes was not statistically significant, FAM-Aβ1-42 internalization was significantly higher in the presence of TTR.

Figure 2: Interaction of FAM-Aβ1-42 with hCMEC/D3 cells in the presence and absence of TTR assessed by flow cytometry:

Figure 2

(A) Internalization levels of FAM-Aβ1-42 by hCMEC/D3 cells in the presence of h rTTR (white columns) was significantly higher than in the absence of the protein (black columns) after 15 and 30 minutes of incubations. (B) Efflux of FAM-Aβ1-42 from hCMEC/D3 measured after 10 minutes of incubation with the peptide was significantly increased at 20 minutes post-replacement with fresh FAM-Aβ1-42-free media, in the presence of h rTTR. N = 3 for each condition and data are expressed as mean±SEM.

Next to investigate the fate of internalized Aβ, we performed an efflux assay. For that, hCMEC/D3 cells were firstly incubated with FAM-Aβ1-42 for 10 minutes, in the absence or presence of h rTTR and then the media were replaced with fresh Aβ-free media. Cells were further incubated at 37 °C and levels of FAM-Aβ1-42 inside cells were measured by flow cytometry, after 10 and 20 minutes. Figure 2B depicts the results showing that in the presence of TTR, FAM-Aβ1-42 effluxes significantly faster than in the absence of this protein, after 20 minutes (45.5% and 67.6% fluorescent cells, respectively).

Effect of TTR in hCMEC/D3 brain-to-blood permeability to Aβ1-42 peptide

In order to investigate the effect of TTR in Aβ1-42 transport across a monolayer of cells, acting as a model of the BBB as previously described, Aβ1-42 transport experiments were performed in hCMEC/D3 cultured in transwells inserts, as shown in Fig. 3A. Cells were grown for 10 days until reaching maximal confluence and allowing TJ formation. Thus, at this point, the cell monolayer should show restricted paracellular permeability, and its confirmation was done using FITC-labelled dextran as a low molecular weight paracellular diffusion marker. In this approach, FITC-labelled dextran 0.25 mg/mL was added to the apical chamber, and then incubated for 1 hour. Wells in which FITC-labelled dextran exceeded 125 ng/mL on the basolateral chamber were considered to have the monolayer disrupted and thus were excluded from the experiment.

Figure 3: Brain-to-blood permeability of hCMEC/D3 cells to Aβ1-42:

Figure 3

(A) Schematic representation of the transwell system used showing the brain and blood sides; Aβ1-42 peptide was always added to the brain side, whereas TTR was added either to the brain or to the blood sides. (B) Brain-to-blood permeability was increased in the presence of h rTTR although without reaching significant differences. However, in the presence of (C) hTTR present in sera, brain-to-blood permeability of hCMEC/D3 cells to Aβ1-42 was significantly increased after 3 hours up to 48 hrs. As a control, Aβ peptide was also added to non-seeded filters to show free passage of the peptide when compared to cell-seeded ones. N = 3 for each condition and data are expressed as mean±SEM. To mimic the absence of TTR, we used TTR-depleted human sera obtained after affinity chromatography, and further analysed by western blot (D) lanes 1- human sera; 2- protein G sepharose beads/anti-human prealbumin antibody; 3-human sera TTR-depleted; 4-Eluted TTR; 5-r hTTR.

We added h rTTR either to the brain or to the blood side, whereas Aβ1-42 was always added to the brain side. Results are displayed in Fig. 3B and show increased permeability of the hCMEC/D3 monolayer to Aβ1-42, when h rTTR is in the brain side, as compared to the levels of Aβ1-42 passage when h rTTR is in the blood side, although the differences were not statistically significant.

To further evaluate the effect of TTR in Aβ1-42 transport across the BBB and in order to obtain a more complex environment in hCMEC/D3 model, we performed the same transwell experiments but using human sera as source of hTTR (TTR concentration 7.5 μg/ml). To mimic the absence of TTR, we used human sera after TTR depletion by affinity chromatography (Fig. 3D). Again, hTTR present in the brain side promoted significant Aβ1-42 transport across the hCMEC/D3, as compared to the situation where hTTR was in the blood side (Fig. 3C). This suggests that TTR participates in Aβ1-42 efflux from the brain through a mechanism that implies TTR/Aβ interaction at the BBB or in its vicinity.

Brain permeability to TTR

Given our evidence in TTR-assisted Aβ transport and to clarify if TTR might be co-transported during such process, we assessed TTR internalization by hCMEC/D3 cells, and as shown in Fig. 4A, TTR was uptaken by these cells.

Figure 4: Permeability of hCMEC/D3 cells to TTR:

Figure 4

(A) hCMEC/D3 cells internalize TTR, as assessed by fluorescence microscopy. (B) hCMEC/D3 cells are permeable to TTR in the brain-to-blood direction but not in the blood-to-brain direction. N = 3 for each condition and data are expressed as mean±SEM.

We next investigated if TTR could cross the hCMEC/D3 monolayer and to assess this, hTTR was added either to the apical or basolateral compartment of the transwells. TTR was then quantified in the media of both chambers and analysed as % TTR that passed to the opposite side. As shown in Fig. 4B, TTR crosses the monolayer in the brain-to-blood direction but not in the blood-to brain direction. This suggests TTR is using a receptor with main expression in the basolateral membrane of the hCMEC/D3 cells.

To confirm these results, we also evaluated TTR clearance in vivo, using TTR−/− mice injected with h rTTR, either intracranially (IC) in the right lateral ventricle or intravenously (IV) in the tail vein. As displayed in Table 1, TTR injected in the brain rapidly reached the periphery as TTR was easily detected in blood, whereas mice injected IV showed negligible levels of the protein in the CSF and brain. Thus, this data corroborates the results obtained in the transwell experiments. This also suggests that TTR can favour Aβ brain efflux but cannot favour its influx, contributing to neuroprotection in AD.

Effect of TTR in Aβ1-42 and Aβ1-40 in AD transgenic mice

Previous work using an AD transgenic model (APPswe/PS1A246E) with different TTR genetic backgrounds (AD/TTR) has demonstrated that Aβ1-42 plasma levels are increased in 7-month old TTR+/− female mice, when compared to TTR+/+ animals11, suggesting a role for TTR in Aβ peripheral clearance.

In this work, to obtain a better knowledge on the effect of TTR in plasma Aβ peptide levels, we extended the study by evaluating not only Aβ1-42 but also Aβ1-40 levels in 3-months old AD/TTR+/+, AD/TTR+/− and AD/TTR−/− female mice. Results are depicted in Fig. 5 and show a negative correlation between TTR and both Aβ1-42 and Aβ1-40. Differences between AD/TTR+/+ and AD/TTR−/− mice were found to be statistical significant for both Aβ peptides. In addition, for Aβ1-42 statistical significant differences were also observed between AD/TTR+/− and AD/TTR−/−.

Figure 5: Effect of TTR genetic reduction in plasma Aβ1-42 and Aβ1-40 levels: Results are shown for 3-month old female mice with three distinct genotypes for TTR: AD/TTR+/+ (N = 5 for Aβ1-42; N = 4 for Aβ1-40), AD/TTR+/− (N = 6 for Aβ1-42; N = 4 for Aβ1-40) and AD/TTR−/− (N = 5 for Aβ1-42; N = 4 for Aβ1-40).

Taken together, our results suggest that TTR influences plasma Aβ by reducing its levels.

Effect of TTR in Aβ1-42 internalization by SAHep cells and primary hepatocytes

Aβ is known to also be delivered at the liver for degradation; therefore, we analysed the effect of TTR in FAM-Aβ1-42 internalization using the SAHep cell line. Uptake of Aβ1-42 peptide increased in the presence of h rTTR showing a positive correlation between Aβ uptake and h rTTR concentration, reaching a maximum of 70% when using 4.5–7.5 μg/mL of TTR in 3 hours (Fig. 6A).

Figure 6: Effect of TTR in Aβ peptide internalization by hepatocytes:

Figure 6

(A) FAM-Aβ1-42 internalization by SAHep cells, in the absence or presence of increasing concentrations of h rTTR, as measured by flow cytometry. TTR concentrations up to 4.5–7.5 μg/mL resulted in increased Aβ internalization by cells. N = 3 for each condition. (B) Flow cytometry of primary cultures of hepatocytes derived from mice with different genetic TTR backgrounds; hepatocytes derived from TTR+/+ mice showed significantly more internalization of FAM-Aβ1-42 than those derived from TTR+/− and from TTR−/−. N =  11, N = 8, N = 14, N = 6 for hepatocytes derived from TTR +/+, TTR +/−, TTR −/− and h rTTR treated TTR −/− mice, respectively. (C) moTTR levels in supernatants of primary hepatocytes measured by ELISA confirmed the genetic reduction in TTR+/− which showed about half of the TTR in TTR+/+, while TTR−/− produced no TTR protein. N = 7 for TTR+/+ and −/− mice and N = 5 for TTR +/−.


To further study the effect of TTR in Aβ1-42 uptake by hepatocytes, and in order to avoid addition of exogenous TTR (since hepatocytes produce TTR), we prepared primary cultures of hepatocytes derived from mice with different TTR genetic backgrounds (TTR+/+, TTR+/− and TTR−/−). TTR secretion was evaluated by ELISA revealing values of approximately 70 and 40 ng/mL for TTR+/+ and TTR+/−, respectively, over a period of 3 hours (Fig. 6C). TTR−/− hepatocytes did not produce TTR, as expected.

As for Aβ1-42 uptake, we observed that TTR facilitated peptide internalization by primary hepatocytes as differences were statistically significant between genetic backgrounds (Fig. 6B). Importantly, addition of h rTTR to TTR−/− hepatocytes partially rescued the phenotype as internalization values equalized those of TTR+/− cells.

Influence of TTR on LRP1 levels

We firstly assessed LRP1 expression by qRT-PCR in total brain extracts of TTR+/+, TTR+/− and TTR−/− mice, and observed significant differences in the expression of this receptor: brains from TTR+/+ mice expressed LRP1 in significantly higher levels than brains from TTR−/− animals (Fig. 7A1). These results were corroborated by measuring LRP1 protein levels by western blot (Fig. 7A2).

Figure 7: LRP1 expression in the brain, liver and cell lines assessed by qRT-PCR, western blot and immunofluorescence: LRP1 levels investigated in the brains from TTR+/+, TTR+/− and TTR−/− mice by

(A1) qRT-PCR (n = 4) and (A2) by western blot (n = 3), showed to correlate directly with TTR levels. hCMEC/D3 cells (n = 3) incubated with TTR showed higher amounts of (B1) mRNA and (B2) protein than cells without TTR. Similarly, livers of TTR+/+ mice expressed more LRP1, both (C1) mRNA (n = 4) and (C2) protein (n = 3), than of TTR−/− mice. (D1) qRT-PCR for LRP1 in SAHep cells incubated with exogenous h rTTR increased their LRP1 mRNA levels (n = 3). (D2) Upon incubation with TTR, SAHep cells increased their LRP1 protein levels.

To further understand the importance of TTR in regulating LRP1 levels in the context of Aβ transport across the BBB, we incubated hCMEC/D3 cells with h rTTR and investigated LRP1 expression by qRT-PCR. As depicted in Fig. 7B1, hCMEC/D3 incubated with TTR displayed higher LRP1 expression, thus confirming the regulation of LRP1 by TTR in these endothelial cells; these results were also corroborated by protein levels, as evaluated by immunocytochemistry (Fig. 7B2)

Similarly to the internalization studies, we also evaluated the ability of TTR to regulate LRP1 levels in hepatocytes by performing qRT-PCR studies in livers from TTR+/+, TTR+/− and TTR−/− mice, as well as in the hepatocyte cell line, SAHep cells. Similarly to the brains, livers from TTR+/+ mice expressed higher levels of LRP1, when compared to the livers from TTR−/− animals (Fig. 7C1). Protein analysis confirmed the effect of TTR at increasing LRP1 and as for the brains, significant differences were observed between TTR+/+ and TTR−/− mice (Fig. 7C2). As for the cell line, SAHep cells analyzed by qRT-PCR (Fig. 7D1) and immunocytochemistry (Fig. 7D2) showed increased LRP1 mRNA and protein levels, respectively, when incubated with TTR.


Altogether, these results indicate that TTR regulates LRP1 levels, suggesting that TTR uses this receptor to promote Aβ clearance.

TTR is a transporter protein mainly synthesized in the liver and in the CP of the brain and secreted into the blood and CSF, respectively. TTR is known to transport several molecules, in particular T4 and retinol through binding to the retinol binding protein (RBP). In the CSF, TTR binds Aβ peptide impeding its deposition in the brain. However, the molecular mechanism underlying this process is not known. Given our earlier evidences that TTR lowers brain and plasma Aβ11, we hypothesized that TTR could function as an Aβ carrier that transports the peptide to its receptor at the brain barriers and at the liver.

Since the cerebral capillaries represent about the double of the total apical surface area of the CP27, we decided to start by studying the effect of TTR in Aβ transport at the BBB. Using the hCMEC/D3 in vitro model of the BBB, we showed that TTR significantly increased Aβ internalization by these cells. Both in the presence and absence of TTR, Aβ internalization levels were high after 15 minutes and no significant increase was measured after 30 minutes. Thus, we assessed efflux by removing media with FAM-Aβ1-42 after a period of incubation to show that TTR was also promoting Aβ efflux from these cells.

To further study the effect of TTR in Aβ transport using the hCMEC/D3 model and given the differential expression of receptors in polarized BBB endothelial cells, we next performed our experiments using transwell cultures. Brain-to-blood transport of Aβ peptide was investigated and we concluded that TTR increased Aβ transport, if added to the brain side but not if added to the blood side. This observation is consistent with a direct TTR/Aβ interaction, as previously demonstrated28. To understand if TTR was also being transported while carrying Aβ, we also evaluated TTR ability to cross the endothelial monolayer to show that this protein can cross in the brain-to-blood direction, but does not cross in the opposite direction. To confirm this, we analyzed in vivo TTR brain permeability using TTR−/− mice injected with h rTTR either into the brain ventricle or into the tail vein. The presence of TTR was then investigated in brain and blood. The results corroborated the in vitroobservations since upon IC administration of TTR, the protein was rapidly found in blood; however, after IV injection of TTR the protein was detected neither in CSF nor in the brain extracts. Our findings are also supported by previous work on TTR turnover and degradation29; in this work authors reported that rat TTR injected intraventricularly into the CSF of rats was mainly degraded in the liver and kidneys (therefore effluxing from the brain), whereas no specific transfer of plasma TTR to the nervous system or degradation of plasma TTR in the nervous system was observed. It is worthy to note that Makover and colleagues injected purified rat TTR in a system containing the same endogenous rat TTR29, and results are similar to the ones we describe now. Therefore, we can conclude that in our system the TTR−/− background did not significantly affected TTR clearance.

The differential brain permeability to TTR indicates the use of a receptor with preferential expression on the basolateral membrane of the endothelial cells forming the BBB, such as LRP1, which in turn is known to internalize Aβ peptide. Whether TTR can cross or not as a complex, namely with Aβ peptide, is not known and needs to be investigated.


TTR gene expression in the brain is usually described as being confined to the CP and meninges, although TTR can be transported to other brain cells. For instance, it is described that in situations of compromised heat-shock response, and as a response to cerebral ischemia, CSF TTR contributes to control neuronal cell death, edema and inflammation12. This implies that TTR is transported from CSF to other brain areas, and thus it is also possible that this protein participates in Aβ transport at the BBB. TTR gene expression has been also attributed to neurons and for instance, SH-SY5Y cells transfected with APP695 isoform showed up-regulation of TTR mRNA expression, with concomitant decrease in Aβ levels16. Other authors showed that the majority of hippocampal neurons from human AD and all those from APP23 mouse brains contain TTR. In addition, quantitative PCR for TTR mRNA and Western blot analysis showed that primary neurons from APP23 mice transcribe TTR mRNA, and that the cells synthesize and secrete TTR protein15. More recently, it has been shown that TTR transcription and protein production can be induced by heat shock factor 1 (HSF1) in hippocampal neurons but not in the liver, both using cell lines and in vivo approaches17.

Importantly, the BCSFB should also be investigated for TTR-assisted Aβ transport, since this protein is the major protein binding Aβ in CSF. In spite of the low TTR levels in CSF (~2 mg/mL), the choroid plexus is presented as the major site of TTR expression, expressed as a ratio of TTR/mass of tissue, corresponding to a ~30-fold higher than that found in plasma30. Interestingly, a recent report describes that in a triple transgenic mouse model of AD only the Aβ1-42 isoform is increased at the epithelial cytosol, and in stroma surrounding choroidal capillaries. Noteworthy, there was increased expression, presumably compensatory, of the choroidal Aβ transporters: LRP1 and RAGE. In addition, authors reported that the expression of TTR was attenuated as compared to non-transgenic mice31.

Previous works indicated that the genetic reduction of TTR in an AD mouse model results in increased Aβ brain levels9,10; another work using 7 month old female mice also showed increased Aβ1-42 plasma levels in AD/TTR+/− mice as compared to age-and gender-matched AD/TTR+/+ animals. In the present work, we extended our study and evaluated both plasma Aβ1-42 and Aβ1-40 isoforms in 3 months old AD/TTR+/+, AD/TTR/+/− and AD/TTR−/− animals, showing that TTR correlates negatively with both isoforms of Aβ. Further, these findings support the idea that plasma may also reflect disease disturbances in AD.

Thus, the following level of our study focused on the effect of TTR in Aβ peptide uptake by the liver. After showing that h rTTR produces a concentration-dependent increase in Aβ internalization by SAHep cells, we worked with primary hepatocytes derived from mice with different TTR backgrounds showing again higher levels of internalization in the presence of TTR.

Interestingly, previous work has shown that TTR is internalized by the liver using a RAP-sensitive receptor20, such as LRP1. Multiple factors influence the function of LRP1-mediated Aβ clearance, such as its expression, shedding, structural modifications and transcriptional regulation by other genes32. Recent studies have clarified how Aβ clearance mechanisms in the CNS are indirectly altered by vascular and metabolism-related genes via the sterol regulatory element binding protein (SREBP2)33. In addition, AD risk genes such as phosphatidylinositol binding clathrin assembly protein (PICALM)34 and apoE isoforms can differentially regulate Aβ clearance from the brain through LRP135.

Consequently, given the importance of this receptor in Aβ clearance both from the brain and at the liver, we evaluated the levels of gene and protein expression in different models. Both LRP1 transcript and protein levels were increased in TTR+/+ brains as compared to TTR−/−. To further confirm the importance of TTR in regulating the levels of LRP1 specifically at the BBB, and contributing to explain the importance of TTR in Aβ clearance, we measured LRP1 in hCMEC/D3 cells with and without incubation with TTR. We observed that the presence of TTR clearly increased the receptor expression, producing significant differences. A similar study was then undertaken for liver and SAHep cells, which again showed regulation of LRP1 expression by TTR. Whether liver TTR regulates liver LRP1 and CSF TTR regulates brain LRP1 is not known and further studies, namely differential silencing of the TTR gene (liver or CP), should be performed.

In a recent study, TTR has been described to regulate insulin-like growth factor receptor I (IGF-IR) expression in mouse hippocampus (but not in choroid plexus) and this effect is due to TTR mainly synthesized by the choroid plexus (and secreted into the CSF) and not by peripheral TTR36. Once more, the possibility for local TTR production has been advanced by some authors16,17, as already mentioned. Finally, it is also known that LRP1 and IGF-IR interact37,38 in a way that the extracellular ligand-binding domain of LRP1 is not involved thus remaining free to bind its ligands. A common link is now established as TTR can regulate the expression of both receptors, albeit in different areas of the brain, opening the possibility for TTR being involved in other processes in the CNS. Moreover, using mice with deleted APP and APLP2, APP has been shown to down-regulate expression of LRP139 via epigenetic events mediated through its intracellular domain (AICD) and to up-regulate TTR, as previously described16. Though it is not known if LRP1 and TTR regulation are part of the same AICD-pathway since TTR levels were not evaluated in the APP and APLP2-deleted mice.

In summary, we show that neuroprotective effects of TTR previously observed in the context of AD are consistent with its role in Aβ clearance at the BBB and liver, and that TTR regulates LRP1 expression, suggesting that TTR is also transported by this receptor. In the future, the TTR-LRP1 cascade should be further investigated for therapeutic targeting.


TTR decreases in the population of both men and women after age 45 years.  This has consequences with respect to AD.  TTR is mainly synthesized by the choroid plexus (and secreted into the CSF) and not by peripheral TTR36, but this declines even earlier than that produced by the liver. (Ingenbleek and Bernstein, 2016).  This suggests a significant role for these age related changes in the development of AD.  Moreover, what has been presented indicates a role for snake venum in increasing the removal of amyloid plaque that develops in AD.  TTR is important in A-beta clearance in liver and BBB.  There was a shift in APP processing towards the amyloidogenic processing in vivo at the end of the 5-month treatment with Aβ1-6A2VTAT(D) that was not observed in shorter treatment schedules with the same compound


MIT scientists find evidence that Alzheimer’s ‘lost memories’ may one day be recoverable    By Ariana Eunjung Cha

Scientists had assumed for a long time that the disease destroys how those memories are encoded and makes them disappear forever. But what if they weren’t actually gone — just inaccessible?

A new paper published Wednesday by the Massachusetts Institute of Technology’s Nobel Prize-winning Susumu Tonegawa provides the first strong evidence of this possibility and raises the hope of future treatments that could reverse some of the ravages of the disease on memory.

“The important point is, this is a proof of concept,” Tonegawa said. “That is, even if a memory seems to be gone, it is still there. It’s a matter of how to retrieve it.”

Zane JaunmuktaneSimon MeadMatthew Ellis, …., A. Sarah WalkerPeter RudgeJohn Collinge & Sebastian Brandner
Nature (10 Sep 2015)

More than two hundred individuals developed Creutzfeldt–Jakob disease (CJD) worldwide as a result of treatment, typically in childhood, with human cadaveric pituitary-derived growth hormone contaminated with prions1, 2. Although such treatment ceased in 1985, iatrogenic CJD (iCJD) continues to emerge because of the prolonged incubation periods seen in human prion infections. Unexpectedly, in an autopsy study of eight individuals with iCJD, aged 36–51 years, in four we found moderate to severe grey matter and vascular amyloid-β (Aβ) pathology. The Aβ deposition in the grey matter was typical of that seen in Alzheimer’s disease and Aβ in the blood vessel walls was characteristic of cerebral amyloid angiopathy3 and did not co-localize with prion protein deposition. None of these patients had pathogenic mutations, APOE ε4 or other high-risk alleles4associated with early-onset Alzheimer’s disease. Examination of a series of 116 patients with other prion diseases from a prospective observational cohort study5 showed minimal or no Aβ pathology in cases of similar age range, or a decade older, without APOE ε4 risk alleles. We also analysed pituitary glands from individuals with Aβ pathology and found marked Aβ deposition in multiple cases. Experimental seeding of Aβ pathology has been previously demonstrated in primates and transgenic mice by central nervous system or peripheral inoculation with Alzheimer’s disease brain homogenate6, 7, 8, 9, 10, 11. The marked deposition of parenchymal and vascular Aβ in these relatively young patients with iCJD, in contrast with other prion disease patients and population controls, is consistent with iatrogenic transmission of Aβ pathology in addition to CJD and suggests that healthy exposed individuals may also be at risk of iatrogenic Alzheimer’s disease and cerebral amyloid angiopathy. These findings should also prompt investigation of whether other known iatrogenic routes of prion transmission may also be relevant to Aβ and other proteopathic seeds associated with neurodegenerative and other human diseases.

The research, described in the journal Nature, involved two groups of mice. One was a normal control and the other was  genetically engineered to have Alzheimer’s-like symptoms. Both groups were given a mild electric shock to their feet. The first group appeared to remember the trauma of the incident by showing fear when placed back in the box where they had been given the shock. The Alzheimer’s mice, on the other hand, seemed to quickly forget what happened and did not have an upset reaction to the box.

Their reaction changed dramatically when the scientists stimulated tagged cells in their brains in the hippocampus — the part of the brain that encodes short-term memories — with a special blue light. When they were put back in the box following the procedure, their memories of the shock appeared to have returned, and they displayed the same fear as their healthy counterparts.

Tonegawa and his colleagues wrote that the treatment appears to have boosted neurons to regrow small buds called dendritic spines that form connections with other cells.


The revelations have “shattered a 20-year paradigm of how we’re thinking about the disease,” Rudy Tanzi, a Harvard neurology professor who is not involved in the research, told the Boston Herald. He said that since the 1980s, researchers believed the memories just weren’t getting stored properly.

The technique used in the study — optical stimulation of brain cells, or “optogenetics” — involves the insertion of a gene into parts of a brain to make them sensitive to blue light and then stimulating them with the light.

In a commentary accompanying the paper, Prerana Shrestha and Eric Klann of the Center for Neural Science at New York University said that the research employed a “clever strategy” and that “the potential to rescue long-term memory in dementia is exciting.”

Doug Brown, director of research at the Alzheimer’s Society, cautioned that the technique is not something that can be translated into a procedure that is safe for the estimated 44 million people worldwide with dementia just yet.

“While interesting,” he told the Guardian, “the practicalities of this approach — using a special blue light to stimulate memory — means that we’re still many years away from knowing if it would be possible to restore lost memories in people.”

Electrical stimulation of the brain may be one alternative scientists can pursue, according to Christine Denny, a neurobiologist at Columbia University. Nature reported that early trials showed that deep-brain stimulation of the hippocampus may improve memory in some Alzheimer’s patients.


Memory retrieval by activating engram cells in mouse models of early Alzheimer’s disease

Dheeraj S. RoyAutumn AronsTeryn I. MitchellMichele PignatelliTomás J. Ryan Susumu Tonegawa

Alzheimer’s disease (AD) is a neurodegenerative disorder characterized by progressive memory decline and subsequent loss of broader cognitive functions1. Memory decline in the early stages of AD is mostly limited to episodic memory, for which the hippocampus has a crucial role2. However, it has been uncertain whether the observed amnesia in the early stages of AD is due to disrupted encoding and consolidation of episodic information, or an impairment in the retrieval of stored memory information. Here we show that in transgenic mouse models of early AD, direct optogenetic activation of hippocampal memory engram cells results in memory retrieval despite the fact that these mice are amnesic in long-term memory tests when natural recall cues are used, revealing a retrieval, rather than a storage impairment. Before amyloid plaque deposition, the amnesia in these mice is age-dependent3, 4, 5, which correlates with a progressive reduction in spine density of hippocampal dentate gyrus engram cells. We show that optogenetic induction of long-term potentiation at perforant path synapses of dentate gyrus engram cells restores both spine density and long-term memory. We also demonstrate that an ablation of dentate gyrus engram cells containing restored spine density prevents the rescue of long-term memory. Thus, selective rescue of spine density in engram cells may lead to an effective strategy for treating memory loss in the early stages of AD.

Figure 1: Optogenetic activation of memory engrams restores fear memory in early AD mice

Optogenetic activation of memory engrams restores fear memory in early AD mice.

ac, Amyloid-β (Aβ) plaques in 9-month-old AD mice (a), in the DG (b), and in the EC (c). d, Plaque counts in HPC sections (n = 4 mice per group). ND, not detected. e, CFC behavioural schedule (n = 10 mice per group). fi, Freezing leve…

Figure 2: Neural correlates of amnesia in early AD mice.close

Neural correlates of amnesia in early AD mice.

a, b, Images showing dendritic spines from DG engram cells of control (a) and AD (b) groups. c, Average spine density showing a decrease in AD mice (n = 7,032 spines) compared with controls (n = 9,437 spines, n = 4 mice per group).


Behavioural rescue and spine restoration by optical LTP is protein-synthesis dependent.

Behavioural rescue and spine restoration by optical LTP is protein-synthesis dependent.

a, Modified behavioural schedule for long-term rescue of memory recall in AD mice in the presence of saline or anisomycin (left). Memory recall 2 days after LTP induction followed by drug administration showed less freezing of AD mice


Turn Off Alzheimer’s Disease

Lomonosov Moscow State University

This image shows the three-dimensional structure of the dimer of the metal-binding domain of beta-amyloid peptide having 'English mutation'. Two peptide molecules connected to each other with the help of zinc ion. Source: This image shows the three-dimensional structure of the dimer of the metal-binding domain of beta-amyloid peptide having 'English mutation'.  Source: Lomonosov Moscow State University

This image shows the three-dimensional structure of the dimer of the metal-binding domain of beta-amyloid peptide having ‘English mutation’. Two peptide molecules connected to each other with the help of zinc ion. Source: This image shows the three-dimensional structure of the dimer of the metal-binding domain of beta-amyloid peptide having ‘English mutation’. Source: Lomonosov Moscow State University

A group of the Lomonosov Moscow State University scientists, together with their colleagues from the Institute of Molecular Biology, Russian Academy of Sciences and the King’s College London, succeeded in sorting out the mechanism of Alzheimer’s disease development and possibly distinguished its key trigger. Their article was published in Scientific Reports.

‘Alzheimer’s disease is a widespread degenerative damage of central nervous system leading to a loss of mental ability.’Until now it was considered incurable,’ tells Vladimir Polshakov, the leading researcher, MSU Faculty of Fundamental Medicine. Though now scientists managed to distinguish the mechanism ‘running’ the disease development, so, a chance appeared to elaborate some new chemical compounds, that may work as an efficient cure.

Several hypotheses are dedicated to the Alzheimer’s disease development. One of the most common is the so-called amyloid hypothesis.

Amyloids (to be precise, beta-amyloid peptides) are molecular constructions of a protein type and in its normal healthy state they provide a protection to the brain cells. They live fast, and having fulfilled their function they fall prey to the work of proteases, the cleaning enzymes that cut all the used protein elements into harmless ‘slags’ that are further reclaimed or removed from a body. However, according to the amyloid hypothesis, at some point something goes wrong, and the cells’ protectors turn to be their killers. Moreover, those peptides start gathering, forming aggregations and hence getting out of the reach of proteases’ cutting blades. Within the amyloid hypothesis this mechanism is more or less precisely described on the later stages of the disease, when the toxic aggregations appeared already and further, when the brain is covered with amyloid plaques. However, the early stage of a beta-amyloid transformation into harmful organic products is highly unexplored.

‘We knew, for example, that a crucial role in initiation of such processes is played by ions of several transition metals, first of all — zinc,’ tells Vladimir Polshakov. ‘Zinc actually conducts a number of useful and healthy functions in a brain, though in this case it was reasonably suspected as a ‘pest’, and particularly as an initiator of a cascade of processes, leading to theAlzheimer’sdisease. However, it remained unclear, what exactly happens during an interaction of zin? ions with peptide molecules, which amino acids bind zinc ions, and how such interaction stipulates a peptide aggregation. We set a goal to clarify at least some of those questions’.

Scientists studied various pathogenic beta-amyloid peptides, their so-called metal binding domains — relatively short peptide regions, capable to bind metal ions. A number of experimental techniques were applied, including nuclear magnetic resonance (NMR) spectroscopy, used to determine the structure of the forming molecular complexes. Some spectra requiring higher sensitivity were additionally measured in London. According to Polshakov, the choice of the studied pathogens was ‘partly a luck’. One of the specimens was the product of so-called ‘English mutation’ — peptide, different from a common beta-amyloid peptide only with one amino acid substitution. Using the NMR spectroscopy scientists managed to sort out chemical processes and structural changes while a peptide molecules interact with zinc ion and undergo further aggregation.

The second pathogen was an isomerized beta-amyloid peptide. It was not different from a normal one in its chemical composition, though one of its amino acid residues, aspartic acid, was in a form with a specific atomic positioning. Such isomerism happens spontaneously, without help of any enzymes, and is related to the ageing processes, another influential factor of the Alzheimer’s disease. Fellow biologists from the Moscow’s Institute of Molecular Biology showed recently, that administration of an isomerized peptide to transgenic mice led to an accelerated formation of amyloid plaques. With the presence of zinc ions, a metal binding domain of the isomerized peptide aggregated so fast that the forming structures were hard to detect. Though scientists managed to distinguish that despite all the differences in processes occurring to the ‘English mutant’ and isomerized peptide in presence of zinc ions, initial stages of these transformations were similar. The trigger happened to be the same — a role of a pathogenic aggregation’s seed was in both cases played by initially formed peptide dimers, i.e. two peptide molecules, connected to each other with help of zinc ion. Such dimers were also detected in normal human peptides, and the difference in all the studied forms could be explained by the speed of formation of corresponding dimer and its proneness to a further aggregation.

Based on their findings, researches proposed the mechanism of zinc-controlled transformation of a peptide-protector into a peptide-killer. That mechanism, scientists notice, explains multiple experimental data, not only gathered by the group, but also collected by their colleagues in other laboratories preoccupied with the Alzheimer’s disease studies. Researchers also hope that thanks to a very certain targeting their discovery would help to produce new medicine capable to block beta-amyloid peptide aggregation stipulated by zinc ions.


Read Full Post »