Shedding Light on the Molecular Pathology of Amyloid Plaques in

Subscriber access provided by BOSTON UNIV

Shedding Light on the Molecular Pathology of Amyloid Plaques in Transgenic Alzheimer’s Disease Mice using Multimodal MALDI Imaging Mass Spectrometry Ibrahim Kaya, Henrik Zetterberg, Kaj Blennow, and Jörg Hanrieder ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.8b00121 • Publication Date (Web): 12 Apr 2018 Downloaded from http://pubs.acs.org on April 12, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Neuroscience

1

Shedding Light on the Molecular Pathology of Amyloid Plaques in

2

Transgenic Alzheimer’s Disease Mice using Multimodal MALDI

3

Imaging Mass Spectrometry

4

Ibrahim Kaya1,2*, Henrik Zetterberg2,3,4,5, Kaj Blennow2,3, Jörg Hanrieder2,3,4,6

5

1

6

405 30 Gothenburg, Sweden

7

2

8

Gothenburg, Mölndal Hospital, House V3, 43180 Mölndal, Sweden

9

3

Department of Chemistry and Molecular Biology, University of Gothenburg, Kemivägen 10,

Department of Psychiatry and Neurochemistry, Sahlgrenska Academy at the University of

Clinical Neurochemistry Laboratory, Sahlgrenska University Hospital, House V3, 43180

10

Mölndal, Sweden

11

4

12

Queen Square, London WC1N 3BG, United Kingdom

13

5

14

Kingdom

15

6

16

412 96 Gothenburg, Sweden

Department of Molecular Neuroscience, Institute of Neurology, University College London,

UK Dementia Research Institute at University College London, London WC1N 3AR, United

Department of Chemistry and Chemical Engineering, Chalmers University of Technology,

17 18 19

*Corresponding Author

20

Ibrahim Kaya

21

Department of Chemistry and Molecular Biology, University of Gothenburg, Analytical

22

Chemistry, Plan 4, Kemivägen 10, SE-412 96, Gothenburg, Sweden

23

[email protected];+46-73-716 86 04

ACS Paragon Plus Environment

ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1

ABSTRACT

2

Senile plaques formed by aggregated amyloid β peptides are one of the major pathological

3

hallmarks of Alzheimer’s disease (AD) which have been suggested to be the primary

4

influence triggering the AD pathogenesis and the rest of the disease process. However,

5

neurotoxic Aβ aggregation and progression are associated with a wide range of enigmatic

6

biochemical, biophysical and genetic processes. MALDI imaging mass spectrometry (IMS) is

7

a label-free method to elucidate the spatial distribution patterns of intact molecules in

8

biological tissue sections. In this communication, we utilized multimodal MALDI-IMS analysis

9

on 18 month-old transgenic AD mice (tgArcSwe) brain tissue sections to enhance molecular

10

information correlated to individual amyloid aggregates on the very same tissue section. Dual

11

polarity MALDI-IMS analysis of lipids on the same pixel points revealed high throughput lipid

12

molecular information including sphingolipids, phospholipids, and lysophospholipids which

13

can be correlated to the ion images of individual amyloid β peptide isoforms at high spatial

14

resolutions (10µm). Further, multivariate image analysis was applied in order to probe the

15

multimodal MALDI-IMS data in an unbiased way which verified the correlative accumulations

16

of lipid species with dual polarity and Aβ peptides. This was followed by the lipid

17

fragmentation obtained directly on plaque aggregates at higher laser pulse energies which

18

provided tandem MS information useful for structural elucidation of several lipid species.

19

Majority of the amyloid plaque-associated alterations of lipid species are for the first time

20

reported here. The significance of this technique is that it allows correlating the biological

21

discussion of all detected plaque-associated molecules to the very same individual amyloid

22

plaques which can give novel insights into the molecular pathology of even a single amyloid

23

plaque microenvironment in a specific brain region. Therefore, this allowed us to interpret the

24

possible roles of lipids and amyloid peptides in amyloid plaque-associated pathological

25

events such as focal demyelination, autophagic/lysosomal dysfunction, astrogliosis,

26

inflammation, oxidative stress and cell death.


Page 2 of 38

Page 3 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

Keywords: Alzheimer’s Disease (AD), MALDI Imaging Mass Spectrometry, Multivariate

2

Image Analysis, Amyloid Plaque Pathology, Aβ Peptides, Sphingolipids, Phospholipids,

3

Lysophospholipids

4

INTRODUCTION

5

Alzheimer’s disease is a fatal neurodegenerative disorder originally characterized by three

6

remarkable neuropathological features including intracellular neurofibrillary tangles (NFTs)

7

consisting of hyperphosphorylated tau protein1, senile plaques formed by aggregated

8

amyloid β peptides (Aβ)2 and lipid granule accumulations (lipoid granules) in neuroglia.3,4

9

According to the amyloid hypothesis5, accumulation of Aβ is the primary influence triggering

10

AD pathogenesis and the rest of the disease process.6 Aβ aggregates into oligomers, which

11

can be of various sizes, and forms diffuse and neuritic plaques in the brain parenchyma,

12

which have been suggested to cause dysfunction and loss of synapses and neurons.7,8

13

However, molecular mechanisms of how monomeric Aβ peptides are converted to neurotoxic

14

oligomers and further to fibrils rich in cross β-sheet motifs are indeterminate.

15

Alois Alzheimer’s originally reported pathologic hallmark of the AD; lipoid granules in

16

neuroglia4 and strong clinical evidence demonstrated in the ensuing years suggest an

17

aberrant lipid homeostasis associated with AD pathology along with peptide and protein-

18

centric mechanisms, in part for the following reasons.3,9 All APP-cleaving secretases, as well

19

as APP, are transmembrane proteins, suggesting a direct relationship of amyloid pathology

20

with neuronal membrane lipids.10 Importantly, growing evidence indicates that amyloidogenic

21

processing of the APP occurs in lipid rafts, largely because functionally active pools of

22

BACE1 and γ-secretase are present in this cholesterol and sphingolipid-rich membrane

23

micro domains.11 It is also known that Aβ exerts its cytotoxic effects at least partly by

24

perturbating cellular membranes, potentially through the modulation of the activity of

25

phospholipases, such as PLA2,12 PLC13 and PLD2.14 On the other hand, lipid species, such as

26

monosialotetrahexosylganglioside (GM1) was found to trigger Aβ generation15 and



1

aggregation.16 Indeed, a major risk factor for the sporadic AD (SAD) is E4 allelic variant of

2

the apolipoprotein (APOE) gene17, which encodes a protein in cholesterol metabolism and

3

lipid transport further linking AD pathology to lipids.18 It is thus clear that there is an

4

interrelating molecular pathology of neuronal lipids, peptides, and proteins associated with

5

AD pathology.9,19,20 Therefore, in situ analysis of plaque-associated both lipidome and

6

peptidome/proteome analysis would provide a basis for further understanding of associated

7

molecular changes in AD pathogenesis.

8

Advances in lipidomics and peptidomics/proteomics, particularly through the use of

9

electrospray ionization mass spectrometry (ESI-MS) has facilitated an accurate profiling of

10

AD-associated changes in lipids such as cholesterol21, plasmalogens22, sphingolipids23, along

11

with amyloid peptides24 and protein species25 such as apoE4 protein26 in brain tissue.

12

However, these studies were performed using brain tissue extracts where spatially confined

13

changes are convoluted, restricting the direct correlation of molecular information with

14

amyloid plaques which are the prominent scenes of AD pathogenesis. Therefore, advanced

15

chemical imaging techniques are required to characterize disease pathology in situ, such as

16

imaging mass spectrometry (IMS).27 This technique is suited to probe amyloid plaque-

17

associated molecular pathology in specific brain regions.28-37 In particular, MALDI-IMS was

18

shown to be a powerful technique to probe amyloid plaque-associated Aβ isoforms and

19

neuronal lipids in transgenic mice and post-mortem human AD brains.29,33-37

20

Although the vast majority of AD cases are classified as late-onset SAD, rare but highly

21

penetrant autosomal mutations in genes encoding both the Aβ precursor protein (APP) and

22

presenilin 1 and presenilin 2, the catalytic components of the γ-secretase, result in early-

23

onset familial AD (FAD)38, sharing certain pathological characteristics with SAD. Therefore,

24

transgenic animal models of the AD can serve as model systems for more thorough

25

investigations of the aggregation mechanisms, cell signaling and metabolic pathways in vivo.

26

For example, the mouse model (tgArcSwe) carrying both the Arctic (E693G) and the


Page 4 of 38

Page 5 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

Swedish (K670M/N671L) mutation displays extensive Aβ deposition with an early onset age

2

of 5–6 months39, and is, therefore, a well-suited model system to study Aβ plaque pathology.

3

In this communication, we attempted to shed a light on the molecular pathology of individual

4

hippocampal Aβ plaques in 18 month-old tgArcSwe mice which can give insights into the

5

pathophysiological processes leading to disease states in the AD. We utilized multimodal

6

MALDI-IMS analysis on the same tissue sections to sketch a high-throughput molecular

7

architecture of individual amyloid plaques. Subsequent high-spatial resolution (10µm) dual

8

polarity lipid and peptide/protein MALDI-IMS analysis on the same regions of single tissue

9

sections allowed correlation of the ion images of lipids including sphingolipids, phospholipids,

10

lysophospholipids with multiple Aβ isoforms at cellular levels in specific brain regions such as

11

hippocampus of tgArcSwe, where extensive amyloid plaque accumulations were observed.

12

Finally, this approach allowed us to discuss the insights into the molecular pathology of even

13

a single amyloid plaque. Therefore, possible roles and interrelations of lipids and Aβ peptides

14

in amyloid plaque-associated pathological events were reviewed and discussed.

15 16

RESULTS AND DISCUSSION

17

High-Resolution, Dual Polarity MALDI-IMS: Enhanced Hippocampal Lipid Spectral

18

Quality and Spatial Molecular Information

19

High-resolution MALDI-IMS is a strong approach to interrogate the spatial distribution profiles

20

of biological molecules associated with spatially-confined histopathological features in

21

neuroscience.36,40-43 1,5-DAN was previously demonstrated to be a highly efficient MALDI

22

matrix via sublimation matrix coating for brain lipids in both negative and positive polarities

23

compared to several commonly used matrix compounds.41 It is also reported that 1,5-DAN

24

gives enhanced spectral quality of brain lipid species in a wide spectral range with few-

25

number of low-energy laser pulses at high spatial resolutions.36 Furthermore, ion signals are

26

distinctly diverse and intense in negative polarity compared to positive polarity36,41 which can



1

be attributed to distinct reductive properties and radical ion transfer abilities of 1,5-DAN44

2

With respect to enhanced correlated lipidome analysis using 1,5-DAN sublimation MALDI-

3

IMS approach, it was also demonstrated that dual polarity lipid analysis can be performed on

4

the same pixel points at high spatial resolutions.37 Therefore, this approach was utilized to

5

reveal a comprehensive lipidome of a hippocampus region of tgArcSwe mice brain. (Figure

6

1) Due to the limitations of the speed of data acquisition in UltrafleXtreme MALDI-TOF/TOF

7

instruments at high spatial resolutions which is restricted by the vacuum stability of 1,5-DAN

8

matrix molecule41, experiments were performed on the hippocampal areas of tgArcSwe

9

coronal brain tissue sections where extensive amyloid plaque aggregates were revealed by

10

previous amyloid specific IHC staining and/or MALDI-IMS studies.33,35,39 Enhanced ionization

11

at high-spatial resolutions using low-energy few-number of laser shots along with the

12

minimized matrix-derived signals revealed a high throughput hippocampal lipid spectral

13

quality in transgenic AD mice (tgArcSwe) brain tissue in both polarities and on the same pixel

14

points. (Figure1)

15

In negative ionization mode, matrix related signals were minimized down to 320 Da. This

16

yielded an enhanced low-molecular-weight lipid signals (320-620 Da) otherwise would be

17

suppressed and/or interfered by matrix-derived signals36, along with the lipids in phospholipid

18

range (620-1000 Da) and relatively high-molecular-weight-species, gangliosides (1000-2000

19

Da). (Figure 1Ia) Therefore, this yielded spatially well-resolved, lipid signals in a wide

20

spectral range as indicated as ion images of various lipid species such as lysophosphatidic

21

acids (LPA 16:0, m/z 409.3), ceramides (Cer 18:0, m/z 564.6), ceramide phosphates (Cer P,

22

m/z 644.6), phosphatidylethanolamines (PE 40:6, m/z 790.6), phosphatidylinositols (PI 38:4,

23

885.6), sulfatides (ST 24:0, 888.6) and gangliosides (GM3 36:1, m/z 1179.7, GM1 36:1, m/z

24

1544.8). (Figure 1Ib) In positive ion mode, matrix related signals were minimized down to

25

420 Da. This allowed us to observe low-molecular-weight lysophosphatidylcholines,

26

otherwise would be suppressed and/or interfered by matrix-derived signals36. (Figure 1IIa)

27

Moreover, intense phospholipid signals, such as phosphatidylcholines, were observed in the


Page 6 of 38

Page 7 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

mass range of 650-900 Da along with the dimer peaks of phospholipids (1400-1700 Da) as

2

previously observed in MALDI-TOF-MS45. (Figure 1IIa) As a result, spatially well-resolved ion

3

images of lysophophatidycholines (LPC 16:0, m/z 496.3, LPC 18:1, m/z 522.3, LPC 18:0,

4

m/z 524.3, LPC 20:4, m/z 544.3) and phosphatidycholines (PC 32:0, m/z 734.6, PC 34:1, m/z

5

760.6, PC 36:4, m/z 782.6, PC 38:6, m/z 806.6) were observed (Figure 1IIb) which can be

6

simultaneously correlated with the ion images obtained in negative polarity on the same pixel

7

points. (Figure 1Ib)



1 2

Figure 1. MALDI-MS spectra acquired subsequently in I a) negative and II a) positive ion mode on

3

the same pixel points from hippocampal region of a coronal tissue section of tgArcSwe mouse brain. I

4

a) Inset spectra in negative ion mode from left to right indicates low molecular weight (320-620 Da)


Page 8 of 38

Page 9 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

lipid species and relatively high-molecular-weight (1000-2000) lipid species, respectively. I b) Ion

2

images of lipid species such as lysophosphatidic acids (LPA 16:0, m/z 409.3), ceramides (Cer 18:0

3

m/z 564.6), ceramide phosphates (CerP 18:0, m/z 644.6), phosphatidylethanolamines (PE 40:6, m/z

4

790.6), phosphatidylinositols (PI 38:4, m/z 885.6), sulfatides (ST 24:0, m/z 888.6) and gangliosides

5

(GM3 36:1, m/z 1179.7, GM1 36:1, m/z 1544.8). II a) Inset spectra in positive ion mode from left to

6

right indicates low-molecular-weight (420-620 Da) lipid species and relatively high-molecular weight

7

(1400-1700) phospholipid dimers, respectively. II b) Ion images of lipid species such as

8

lysophosphatidylcholines (LPC 16:0, m/z 496.3, LPC 18:1, m/z 522.3, LPC 18:0, m/z 524.3, LPC 20:4,

9

m/z 544.3) and phosphatidycholines (PC 32:0, m/z 734.6, PC 34:1, m/z 760.6, PC 36:4, m/z 782.6, PC

10

38:6, m/z 806.6). Scale bar: 250 µm.

11 12

On the other hand, enhanced ionization of lipid species at high-spatial resolutions allowed us

13

to observe pathological lipid changes not only brain region-specific but also directly

14

correlated to the plaque aggregates (Figure 2), whose amyloid β identities were

15

demonstrated by subsequent peptide imaging on the same imaging sequence in the

16

hippocampus of tgArcSwe mice brain. (Figure 3) To further demonstrate the relevance of

17

high spatial resolution to probe plaque-associated lipid changes, we compared ion images

18

obtained at 30µm and 10µm spatial resolutions in equivalent hippocampal regions of

19

subsequent tissue sections. (Figure 2) For example, ion images of ceramide phosphates

20

(CerP

21

phosphatidylinositols (PI 38:4, m/z 885.6) display distinct amyloid-plaque associated

22

localizations which can be observed explicitly at a spatial resolution of 10 µm (Figure 2b) but

23

not conclusively with 30µm (Figure 2a). In detail, accumulation of CerP (18:0, m/z 644.6) and

24

PI (38:4, m/z 885.6) in the core and the periphery of certain amyloid plaques, respectively,

25

along with the mild reduction of PE (40:6, m/z 790.6) in the certain amyloid plaques in

26

spatially well-resolved hippocampal regions can be luculently observed at 10µm spatial

27

resolutions. (Figure 2b)

18:0,

m/z

644.6),

phosphatidylethanolamines

(PE


40:6,

m/z

790.6),

and


1 2

Figure 2. MALDI-IMS analysis of tgArcSwe mouse hippocampal lipids of coronal mice brain tissue

3

section. Ion images of ceramide phosphates (CerP 18:0, m/z 644.6), phosphatidylethanolamines (PE

4

40:6, m/z 790.6), phosphatidylinositols (PI 38:4, 885.6) were obtained in negative ion mode at spatial

5

resolutions of a) 30 µm, b) 10 µm on the consecutive brain tissue sections. Scale bars: 2 mm.

6 7

Amyloid Plaque Identification by High Spatial Resolution Peptide MALDI-IMS

8

Subsequent to Dual Polarity Lipid MALDI-IMS on the Same Tissue Section

9

1,5-DAN via sublimation allows enhanced ionization of lipid species on tissue sections

10

without the need of high laser pulse energies.36 This ‘’gentle’’ MALDI-IMS methodology

11

preserves the tissue morphology by minimizing the thermal and mechanical nanosecond

12

laser ablation effects during the laser desorption/ionization process36 and therefore allows

13

successful fluorescent immunohistochemistry (IHC)36 or peptide/protein MALDI-IMS37 on the

14

same tissue section.

15

Amyloid identity and fibrillary structure of plaques in tgArcSwe mouse brain tissue sections

16

were previously demonstrated subsequent to MALDI-IMS using a monoclonal Aβ antibody

17

(Aβ1-16, 6E10) and a fluorescent amyloid fibril probe (h-FTAA), a luminescent conjugated


Page 10 of 38

Page 11 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

oligothiophene (LCO)46, respectively.35 However, antibody-based approaches and amyloid

2

fibril-specific staining techniques are limited with respect to chemical specificity. For example,

3

IHC allows simultaneous detection of one or few amyloid isoforms with questionable

4

specificity while fluorescent amyloid fibril staining reveals only structural information of

5

amyloid plaques.35,36 Further, IHC staining subsequent to MALDI-IMS analysis of lipids might

6

be inconclusive due to the thermal and/or mechanical damage on the protein epitopes during

7

laser desorption process.33,36 Alternatively, in order to demonstrate the multiple amyloid β

8

identity of plaque aggregates in the same hippocampal regions (Figure 2), subsequent

9

MALDI-IMS can be applied for peptide/protein analysis subsequent to lipid MALDI-IMS on

10

the same region of the same tissue section.37 Therefore, we employed tri-modal MALDI-IMS

11

paradigm for negative- and positive ion mode lipid analysis and subsequent peptide/protein

12

ion imaging on the same regions at 30µm and 10µm spatial resolutions on the consecutive

13

tissue sections. (Figure 3) High spatial resolution (10 µm) tri-modal MALDI-IMS revealed an

14

accurate correlation of dual polarity lipid ion images such as phosphatidylinositols (PI 38:4,

15

m/z 885.6) in negative ion mode and lysophosphatidylcholines (LPC 16:0, m/z 496.3) in

16

positive ion mode with the ion image of amyloid peptides such as the prominent Aβ 1-40

17

(Figure 3b), whereas the correlation was not conclusive at 30 µm spatial resolution. (Figure

18

3a) These results are potentially important as this high spatial resolution method allows

19

correlation of tri-modal ion images to not only brain regions but also to individual plaque

20

features which are spatially well-resolved. (Figure 3b)



1 2

Figure 3. Trimodal MALDI-IMS analysis of tgArcSwe mouse hippocampal lipids and peptides on the

3

same tissue section. Ion images of phosphatidylinositols (PI 38:4, m/z 885.6) in reflective negative ion

4

mode, lysophosphatidylcholines (LPC 16:0, m/z 496.3) in reflective positive ion mode and amyloid

5

peptide (Aβ 1-40, m/z 4257.6) in linear positive ion mode were obtained at a) 30µm and b) 10µm

6

spatial resolutions on consecutive brain tissue sections. Scale bars: 2 mm.

7

All in all, these methodologies suggest a high throughput investigation of molecular lipid

8

microenvironment of individual Aβ plaques and elucidate their plaque-associated alterations.

9

Simultaneous detection of lipids, irrespective of their ionization polarity, and amyloid peptides

10

correlated to single plaque features allows the discussion of their interrelated amyloid

11

plaque-associated molecular pathology. Although tgArcSwe mice, as well as the other

12

transgenic mice species, do not fully represent the human AD disease47, they can provide

13

valuable insights into disease mechanisms. In this regard, senile plaques in human AD

14

brains are mainly smaller (10-20µm)48 as well as having different chemical composition49,50,

15

physicochemical properties48 and diverse structural heterogeneity50,51 compared to the

16

plaques in transgenic AD mice brains. Therefore, this methodology might require

17

adjustments for human AD brain tissue applications. On the other hand, it can be also

18

restricted due to the limit of spatial resolution can be achieved using UltrafleXtreme MALDI-


Page 12 of 38

Page 13 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

TOF/TOF instruments (5-10µm). However, recent advancements in the instrumentation and

2

laser focusing technologies allow sub-cellular spatial resolution (~1µm) MALDI-IMS

3

experiments on brain tissue sections.52,53

4 5

Multivariate Image Analyses Reveal Correlation of Plaque-Associated Lipid and

6

Amyloid Peptide Species

7

In order to probe the complex multimodal imaging data in an unbiased way, multivariate

8

image analyses were performed. The acquired image data were investigated by using

9

unsupervised multivariate statistics based on hierarchical clustering analysis (HCA, bisecting

10

k-means) in order to obtain image segmentation of anatomical regions of interest based on

11

their lipid chemical profile. Image segmentation identified deposit-like features in the

12

hippocampus (Figure 4, (Supporting Information Figure S-1 and S-2). These features were

13

assigned as individual regions of interest (ROI) and correlated to the processed MS data in

14

order to identify the associated chemical species that allowed image segmentation of the

15

regions. For positive ion-mode IMS of lipids, no conclusive localization that resembles plaque

16

pathology was observed with cluster analysis, possibly due to the large number of

17

phospholipids that display an unspecific distribution and convolute lipid species that localize

18

to plaques using this kind of data analysis. In order to validate the HCA analysis and expand

19

the MVA strategy, an additional principal component analysis was performed for the trimodal

20

IMS dataset (Figure 4, Supporting Information Figure S-3). Similar to the HCA results, plaque

21

features were distinct as demonstrated by the corresponding score images (Figure 4). In

22

contrast, PCA allowed identifying plaque-associated lipids detected in positive ion mode

23

(Figure 4h), that however showed only very little contribution to overall variance

24

demonstrating the sensitivity of that approach.

25

Further, inspection of the corresponding HCA and PCA loadings provided insight in the

26

underlying chemical differences and PCA revealed a number of variables (m/z values) that



1

were annotated as distinct lipid or peptide species based on accurate mass that displayed

2

spatial, plaque-associated correlation.

3 4

Figure 4. Multivariate image analyses of trimodal imaging MALDI-MS data. a) Brightfield image of

5

hippocampal area analyzed by IMS3. Hierarchical clustering and PCA of IMS protein (b-d), lipid data

6

acquired in negative ion mode (e,f) and lipids acquired in positive ion mode (g,h). (a) HCA

7

segmentation images reveal plaque features for peptide/protein data (c, b: semitransparent overlay

8

with brightfield image and negative ion mode lipid IMS data (e) and matching PCA score images (d,

9

proteins: PC2; f, lipids neg. PC5) show identification of plaque features. For positive ion mode lipid

10

IMS, HCA was less conclusive while PCA (h, PC7) allows outlining plaque-associated lipids. Scale

11

bar: 250 µm.

12 13

Amyloid Plaque-Associated Alterations of Lipids in tgArcSwe Mice Brain Revealed by

14

Multimodal MALDI-IMS: Insights Into the Molecular Pathology

15

Amyloid plaque formation is a dynamic process which is associated with several enigmatic

16

biochemical, biophysical and genetic processes. Most of the lipid classes were suggested to

17

be crucial players in AD pathogenesis including amyloid plaque pathology9,20 Based on the

18

published studies, one could have in view following main mechanisms for the action of lipids


Page 14 of 38

Page 15 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

in amyloid pathology and senile plaque formation: firstly, the interaction of amyloid peptides

2

and/or amyloid oligomers with the cellular membranes defecting the membrane lipid

3

homeostasis and/or altering the organization and physicochemical properties of

4

membrane bilayer which potentially alters APP processing.13,15,54,55 Secondly, the functional

5

roles of peptides/proteins such as Aβ and APP Intracellular domain (AICD) in the regulation

6

of lipid metabolism pathways either directly or by affecting the gene expressions encoding for

7

APP, BACE1, the Aβ-degrading protease neprilysin (NEP) as well as several enzymes

8

involved in lipid metabolism.20,56-60 Bidirectionally, lipid-peptide/protein interactions, whereby

9

changes in lipids modulate the function of target proteins such as APP and APP-cleaving

10

secretases.61,62 Further, catalyzer and/or interfacial role of lipids in amyloid oligomerization

11

and fibrillization.16,63,64 Last but not least, differential roles of Apo-E isoforms in Aβ

12

aggregation and clearance along with their distinct roles in regulating CNS lipid transport.65,66

13

Considering the heterogeneity of lipid molecules and complex bidirectional link between

14

lipids and amyloid pathology, it is expectative to observe amyloid plaque-associated

15

alterations of several lipid classes in tgArcSwe mice brain tissue. Here, high spatial

16

resolution, high throughput MALDI-IMS and multivariate image analysis reveal amyloid

17

plaque-associated alterations of several lipids including sphingolipids, phospholipids and

18

lysophospholipids, which can give insights in to the molecular pathology of AD pathogenesis.

the

19 20

Sphingolipids including ceramides, sphingomyelins, and glycosphingolipids (GSLs) are major

21

components of lipid rafts, playing a number of crucial parts in cell functions.67 Growing

22

evidence indicating that the amyloidogenic processing of APP occurs in lipid rafts11 closely

23

links sphingolipids to amyloid pathology. Moreover, the APP-processing product AICD

24

downregulates the gene expression encoding for serine palmitoyl-CoA (SPT) enzyme which

25

initiates the sphingolipid biosynthesis.57 In addition, a number of studies have been reported

26

on the alterations of sphingolipids in post-mortem human and transgenic mice AD-brain

27

tissue and their potential to modulate APP processing15,68 and Aβ aggregation.69,70



1

High-spatial-resolution MALDI-IMS was performed on hippocampal and adjacent cortical

2

regions of a coronal tgArcSwe mice brain tissue section. (Figure 5) Inspection of ion images

3

in reflective negative ion mode revealed plaque-associated alterations of sphingolipids such

4

as the accumulation of ceramides (Cer 14:0, m/z 508.6, Cer 18:0, m/z 564.6), and

5

gangliosides (GM3 36:1, m/z 1179.8, GM3 38:1, m/z 1207.8, GM2 36:1, m/z 1382.8 ) along

6

with the depletion of sulfatides (ST 18:0, m/z 806.8, ST 22:0, m/z 862.8, ST 22:0 (OH), m/z

7

878.8, ST 24:0, m/z 890.8, ST 24:0 (OH), m/z 906.8). Further, subsequent MALDI-IMS for

8

peptides was performed on the same region of the tissue section to identify the amyloid

9

deposits. An intense accumulation of Aβ 1-40, m/z 4257.6 (Figure 5m) along with the other

10

amyloid isoforms (Aβ 1-37, m/z 4002.7, Aβ 1-38, m/z 4060.3 and Aβ 1-42, m/z 4441.2)

11

(Figure 5p) all colocalized with spatial lipid alterations.

12

Amyloid plaque-associated alterations of certain sphingolipids were previously described for

13

the same transgenic mice (tgArcSwe) using MALDI-IMS and subsequent IHC and

14

fluorescent staining on the same tissue section.35 Here, multimodal MALDI-IMS reveals

15

amyloid-plaque associated alterations of several sphingolipid species all correlated to

16

multiple Aβ isoforms at 10µm spatial resolution. (Figure 4) Previously, shotgun lipidomics

17

consistently revealed the substantial decrease of sulfatides and the large content increase of

18

ceramides in post-mortem human AD brain extracts relative to cognitively normal age-

19

matched controls.23 Furthermore, a recent study performed with MALDI-IMS indicated

20

regional depletion of cortical sulfatides in post-mortem human AD-brain tissues from pre-

21

clinical to severe stages of the AD.71 Our data indicates both hydroxylated and non-

22

hydroxylated sulfatides, irrespective of their fatty acid (FA) moiety, depletes on the amyloid

23

plaques (Figure 4e-i). Interestingly, ST (18:0, m/z 806.8) species (abundant in neurons and

24

astrocytes)72 shows lesser depletion relative to the long-chain sulfatides (enriched in mature

25

oligodendrocytes)73 in the center of the plaques, whereas ST (18:0, m/z 806.8) species also

26

reflected a clear elevation on the periphery of the plaques. (Figure 5e) Further, elevation of

27

ceramides (Cer 14:0, m/z 508.6, Cer 18:0, m/z 564.6) were correlated with the depletion of

28

all the sulfatides (Figure 5n) and deposition of Aβ peptide isoforms (Figure 5m, 5p). On the


Page 16 of 38

Page 17 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

other hand, ceramide phosphates (CerP 18:0, m/z 644.6) reflected an inconsistent plaque-

2

associated distribution. They were elevated in the cores and depleted on the edges of certain

3

plaques (Figure 2b, Figure 5d) which might be attributed to the structural heterogeneity

4

and/or dynamics of plaque generation.

5 6

Figure 5. Amyloid plaque-associated alterations of sphingolipids correlated with amyloid peptide

7

isoforms revealed by Multimodal MALDI-IMS on the same tissue section of tgArcSwe mouse brain.a)



1

brightfield image of coronal brain tissue section of tgArcSwe b) Cer 14:0, m/z 508.6, c) Cer 18:0, m/z

2

564.6, d) CerP 18:0, m/z 644.6, e) ST 18:0, m/z 806.8, f) ST 22:0, m/z 862.8, g) ST 22:0 (OH), m/z

3

878.8, h) ST 24:0, m/z 890.8, i) ST 24:0 (OH), m/z 906.8, j) GM3 36:1, m/z 1179.8, k) GM3 38:1, m/z

4

1207.8, l) GM2 36:1, m/z 1382.8, m) Aβ 1-40, m/z 4257.6, n) merged ion images of sulfatides (green)

5

and ceramides (red), o) merged ion images of sulfatides (red) and gangliosides (green), p) merged ion

6

images of sulfatides (green) and amyloid peptides isoforms (blue) (Aβ 1-37, m/z 4002.7, Aβ 1-38, m/z

7

4060.3, Aβ 1-40, m/z 4257.6 and Aβ 1-42, m/z 4441.2). Scale bar: 2mm.

8 9

Sulfatides are one of the main consituents of the myelin sheath surrounding axons in the

10

CNS and mainly synthesized by oligodendrocytes playing key roles in the regulation of

11

oligodendrocyte maturation and myelin formation.73,74 Ceramides are the central components

12

in sphingolipid metabolism and serve as the backbones to generate sphingomyelin (SM) and

13

GSLs including sulfatides and gangliosides. Early reports indicated the role of ApoE in both

14

CNS and peripheral nervous system which is the regulation of sulfatide level.65 Further, it has

15

been hypothesized that the alterations of apolipoprotein E (APOE) ε4 allele-mediated

16

sphingolipid homeostasis through lipoprotein pathways can lead the depletion of sulfatides in

17

myelin sheaths.75,76 On the other hand, the ceramide elevation in AD was suggested to be a

18

possible degradation product of sulfatides23, but other explanations include Aβ induced

19

membrane oxidative stress77, Aβ-mediated activation of sphingomyelinases (SMases)

20

catalyzing the breakdown of SM to ceramide78,79, and multiple gene expression abnormalities

21

such as increased cerebral expression of genes involved in ceramide de novo synthesis

22

along with the reduced expression of genes reguired for glycosphingolipid synthesis out of

23

ceramide.80 Indeed, oligodendrocytes was suggested be highly prone to be affected by

24

oxidative stress possibly due to the required energy for the production of lipid-rich and

25

protein dense myelin sheath.81,82 Further, amyloid peptides such as Aβ 1-40 and Aβ 1-42

26

were shown to be cytotoxic83,84 and Aβ 1-40 can induce the death of oligodendrocytes by

27

activating sphingomyelinase-ceramide pathway resulting in increased ceramide generation.79

28

In line with this, ceramides are pro-apoptotic and neurotoxic signaling molecules85,86 and they


Page 18 of 38

Page 19 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

were associated with dysregulated apoptosis in AD.77 On the other hand, Cer P species

2

produced through ceramide kinase (CERK) mediated phosphorylation were described as a

3

potent inhibitor of cell death and inducer of cell survival.87,88 Therefore, phosphorylation of

4

ceramides in the cores of certain plaques (Figure 2b, Figure 5d) might suggest cellular

5

defense mechanism during Aβ plaque maturation, whereas ceramides indicate excessive

6

amyloid-associated cell death. (Figure 5b-c) Our data suggest both sulfatide depletion and

7

ceramide elevation are amyloid plaque-associated phenomena in tgArcSwe AD mice

8

pathology (Figure 5m,n,p). Indeed, earlier studies performed by quantitative fluorescent

9

staining revealed amyloid plaque-associated demyelination and oligodendrocyte loss in both

10

human and transgenic mice AD-brain.89,90 Therefore, substantial amyloid-plaque associated

11

depletion of sulfatides could reinforce the previous results indicating amyloid plaques could

12

be associated with regression of myelination which is another pathological hallmark of AD.91

13

92

14

Along with sulfatide depletion and ceramide elevation, substantial accumulation of

15

gangliosides (GM3 36:1, m/z 1179.8, GM3 38:1, m/z 1207.8, GM2 36:1, m/z 1382.8) were

16

revealed on the same amyloid plaques. (Figure 5j-l, o) A majority of the earlier studies

17

indicated a general reduction of gangliosides93-95 along with the elevation of simpler

18

gangliosides (GM2, GM3)95,96 in both transgenic mice and human AD brain extracts which

19

was hypothesized to be a result of accelerated lysosomal degradation and/or astrogliosis

20

during neuronal cell death.96 This in line with the reported accumulations of GM2 and GM3 in

21

lysosomal storage disorders.97,98 Astrocytes, the most abundant glial cell type in the nervous

22

system, respond to neurodegenerative disorders including AD through astrogliosis, i.e.,

23

converting to a reactive inflammatory state.99 Indeed, a recent study revealed the connection

24

of astrogliosis with Aβ plaque formation through probing vimentin and GFAP (biomarkers of

25

astrogliosis) in tgArcSwe mice pathology which reinforces this hypothesis.100 Concerning this,

26

elevation of ST (18:0) species, abundant in astrocytes,72 on the periphery of amyloid plaques

27

might indicate amyloid plaque-associated astrogliosis in tgArcSwe mice. (Figure 5e) Further,

28

gene expression studies in post-mortem human brains also revealed defects in the



1

lysosomal system in association with AD.101,102 It is also evidenced that autophagic/lysosomal

2

system is dysfunctioned in the AD in association with peptide/protein accumulation.103,104

3

Autophagy is a natural cellular pathway involved in the turnover of proteins, protein

4

aggregates, and damaged organelles as well as intracellular pathogens through lysosomal

5

degradation.105 In the AD, it was suggested that the maturation of autophagolysosomes and

6

their retrograde transport are disrupted, which results in excessive accumulation of

7

autophagic vacuoles (AVs) within degenerating neurites. This combination of increased

8

autophagy induction and incomplete clearance of Aβ-generating AVs likely results in Aβ

9

accumulation.106,107 Therefore, accumulation of simpler gangliosides (GM2, GM3) on the Aβ

10

peptide aggregates might be a result of amyloid plaque-associated autophagic/lysosomal

11

dysfunction in tgArcSwe mice pathology. (Figure 5j-l, o)

12

Phospholipids are essential constituents of neuronal membranes providing structural integrity

13

and functional properties.108 Lysophospholipids, metabolic intermediates of phospholipid

14

metabolism109 interact with the lipid and protein moieties of neural membranes and modulate

15

the function of neural membrane proteins, such as enzymes and growth factors.110 In

16

addition to the sphingolipids, alterations in the levels of phospholipids, phosphoinositides

17

(phosphorylated derivatives of phospholipids) and lysophospholipids along with the

18

phospholipid-metabolizing enzymes such as PLA2 and PLC, and phosphoinositide kinases

19

and phosphatases as well as their potential roles in processes regulating Aβ generation such

20

as modulation of APP111 and presenilin112 were reported in post-mortem human and

21

transgenic AD mice brain tissues.113-117

22

Despite the high abundance and the efficient ionization of many of them, the structural

23

diversity of phospholipids and lysophospholipids limit their comprehensive screening within a

24

single MALDI-IMS analysis. Therefore, subsequent dual polarity lipid MALDI-IMS on the

25

same tissue section increases the number of detected lipid species which can be correlated

26

to spatial histopathological features.37,41,118 Here, high spatial resolution, dual polarity

27

analysis on the hippocampus of tgArcSwe mice brain tissue allowed us to simultaneously

28

correlate the alterations of phospholipid (Figure 6) and lysophospholipid (Figure 7) species,


Page 20 of 38

Page 21 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

irrespective of their ionization polarity, to microscopic amyloid plaque structures. In detail,

2

lysophosphatidic acids (LPA), phosphatidic acids (PA), lysophosphatidylethanolamine (LPE),

3

phosphatidylethanolamines (PE), phosphatidylserines (PS), lysophosphatidylinositols (LPI)

4

and phosphatidylinositols (PI)

5

lysophosphatidylcholines (LPC) and phosphatidylcholines (PC) which are observed in

6

positive ion mode on the same pixel points (Figure 6,7). Inspection of ion images in

7

phsopholipid region of the spectra revealed depletion of PE (40:6, m/z 790.6), PE-p (40:6,

8

m/z 774.6), PS (40:6, m/z 834.6) species along with the elevation of PA (32:0, m/z 647.5),

9

PE-NMe2 (32:0, m/z 718.5), PC (32:0, m/z 734.6), PE (36:1, 744.5), PI (36:4, m/z 857.6), PI

10

(38:5, m/z 883.6), PI (38:4, m/z 885.6) species on the amyloid plaque regions (Figure 6).

11

Interestingly, PA (34:2 m/z 671.5) species revealed interesting distributions. They were

12

elevated in the cores and depleted on the edges of certain plaques which is similar to CerP (

13

18:0, 644.6) species (Figure 6c).

are observed in negative ion mode can be correlated

14 15

Figure 6. Amyloid plaque-associated alterations of phospholipids including phosphatidic acids (PA),

16

phosphatidylethanolamines

(PE),

phosphatidylcholines

(PC),


phosphatidylserines

(PS)

and


1

phosphatidylinositols (PI) were revealed by dual polarity MALDI-IMS on the same pixel points. a)

2

brightfield image of coronal brain tissue section of tgArcSwe, b) PA (32:0, m/z 647.5), c) PA (34:2 m/z

3

671.5), d) PE-NMe2 (32:0 m/z 718.5 ), e) PC (32:0 m/z 734.6), f) PE (36:1 744.5), g) PE (40:6 m/z

4

790.6), h) PE-p (40:6 m/z 774.6), i) PS (40:6 m/z 834.6), j) PI (36:4 m/z 857.6), k) PI (38:5 m/z 883.6)

5

and l) PI (38:4 m/z 885.6). Scale bar: 2 mm.

6 7

Early reports generally indicated reductions of PE-PLs, PC-PLs, and PI-PLs in human post-

8

mortem and transgenic AD-brains22,113,119,120 which was speculated to be due to the increased

9

oxidative stress and stimulated activity of phospholipases in the presence of Aβ peptides

10

resulting in phospholipid (PL) degradation.116,121,122 Our data indicates a clear depletion of

11

phospholipids containing polyunsaturated fatty acids (PUFA) in their structure including PE

12

(40:6 m/z 790.6), PE-p (40:6 m/z 774.6), PS (40:6 m/z 834.6) on the amyloid plaques which

13

might be a result of lipid peroxidation by amyloid plaque-associated oxidative stress as

14

previously demonstrated122-124 (Figure 6g,h,i). Indeed, a previous study suggested that Aβ-

15

induced reactive oxygen species (ROS) destabilizes the alkyl-dihydroxyacetone phosphate

16

synthase (AGPS) protein which is responsible for the initial de novo synthesis reaction of

17

plasmalogens125, the common phospholipids enriched in PUFA.58 On the other hand, PA

18

(32:0, m/z 647.5), PE-NMe2 (32:0, m/z 718.5), PC (32:0, m/z 734.6), PE (36:1, 744.5)

19

species containing no PUFA in their structure were elevated on amyloid plaques which can

20

be explained by the possible interfacial and/or catalyzer roles of these lipids in amyloid

21

oligomerization/aggregation126 (Figure 6b,d,e,f). Deposition of PA (34:2, m/z 671.5) species

22

in the center of plaques might suggest an early hyperactivity of PLD2, which hydrolyzes PC to

23

PA114 (Figure 6c). Indeed, an early study indicated that swAPP transgenic mice lacking PLD2

24

are protected from the synaptotoxic and memory-impairing actions of Aβ in vitro and in vivo14

25

along with normalizing GM3 levels in the brain.127 Therefore, this might suggest that the early

26

form of Aβ, possibly Aβ oligomers, can stimulate the activity of PLD2 which is halted after

27

plaque maturation.


Page 22 of 38

Page 23 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

In contrast to earlier reports on brain tissue extracts113, phosphatidylinositol including PI

2

(36:4, m/z 857.6), PI (38:5, m/z 883.6), PI (38:4, m/z 885.6) were elevated and localized to

3

amyloid plaques, irrespective of their FA moiety (Figure 6j,k,l). While this can be due to the

4

pathogenesis difference between human and transgenic mice AD-brains, our results

5

represent the alterations of this species in spatially confined amyloid plaques. Inositol

6

phospholipids and phosphoinositides have been known to be important regulatory molecules

7

in cell regulation and membrane dynamics.128 For example, phosphatidylinositol-3-phosphate

8

(PI3P), a phosphoinositide synthesized primarily by lipid kinase Vps34 plays essential roles

9

in autophagy along with the sorting and processing of APP through the endosomal system.129

10

It was evidenced that disruption of Vps34 impairs autophagy, lysosomal degradation as well

11

as lipid metabolism and enhances amyloidogenic processing of APP.129,130 In AD pathology,

12

reduction of phosphatidylinositol kinase activity and depletion of phosphoinositide species in

13

human AD-brain tissue were previously reported.113,117 Furthermore, oligomeric amyloid-β

14

peptide was demonstrated to hinder phosphatidylinositol-4,5-bisphosphate (PtdIns(4,5)P2)

15

levels in neurons.13 Therefore, elevation of phosphatidylinositols on amyloid plaques might

16

be associated with disruption of phosphoinositide formation pathways which can be

17

associated autophagic/lysosomal dysfunction and defected lipid metabolism such as

18

sphingolipids130 (Figure 5). In addition to this, interaction of PI species with Aβ 1-40 and Aβ 1-

19

42 was shown to act as a seed for amyloid fibril formation64 in vivo suggesting another

20

possible role of PIs in association with plaques.

21

Along with the amyloid plaque-associated alterations of phospholipids, inspection of ion

22

images revealed accumulation of lysophospholipids including LPA (16:0, m/z 409.3), LPA (

23

18:1, m/z 435.3), LPA (18:0, m/z 437.3), ) LPE (16:0, m/z 452.5), LPE (18:0, m/z 480.5), LPC

24

(16:0, m/z 496.3), LPC (18:1, m/z 522.3), LPC (18:0, 524.3) to amyloid plaques (Figure 7).

25

Elevation of LPC and depletion of PC species were previously demonstrated in the brains of

26

9 month-old transgenic AD mice (5XFAD) using MALDI-IMS.116 Further, immunostaining of

27

amyloid peptides and PLA2 in 5XFAD mice indicated an extensive accumulation of amyloid

28

peptides in the cortex and subiculum along with the partial accumulation of PLA2 around the



Page 24 of 38

1

Aβ plaques.116 Therefore, this suggests an amyloid plaque-associated hyperactivity of PLA2

2

which can result in the accumulation of lysophospholipids.131 Indeed, increased neuronal

3

activity of PLA2 converts PC to LPC, which stimulates glial cells and subsequently induces

4

astrogliosis, neuroinflammation and triggers neurodegeneration.132 In line with this, activation

5

of PLA2 was also observed where inflammation present in post-mortem human AD brains in

6

association with Aβ pathology,133 Regarding tgArcSwe mice brain, Aβ plaque formation-

7

associated astrogliosis were observed which can also trigger inflammatory PLA2 activity.100

8

Our data reveals that lysophospholipid accumulation is associated with amyloid plaques

9

along with the alterations of sphingolipids and phospholipids in tgArcSwe AD mice pathology.

10 11

Figure 7. Amyloid plaque-associated alterations of lysophospholipids including lysophosphatidic

12

acids

13

lysophosphatidyinositols (LPI) were revealed by dual polarity MALDI-IMS on the same pixel points. a)

14

brighfield image of the coronal brain tissue section of tgArcSwe. b) LPA (16:0, m/z 409.3), c) LPA (

15

18:1, m/z 435.3), d) LPA (18:0, m/z 437.3), e) LPE (16:0, m/z 452.5), f) LPE (18:0, m/z 480.5), g) LPC

16

(16:0, m/z 496.3), h) LPC (18:1 m/z 522.3), i) LPC (18:0 524.3), j) LPC (20:4, m/z 544.3), k) LPI (16:0,

17

571.3), l) LPI (18:0, m/z 599.3). Scale bar: 2mm

(LPA),

lysophoshatidylethanolamines

(LPE),

lysophosphatidylcholines


(LPC)

and

Page 25 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

Apart from being the possible metabolites of amyloid plaque-associated phospholipase

2

hyperactivity, deposition of LPC species on the plaques can be attributed to their roles in

3

amyloid oligomerization63 and amyloid fibrillization134 along with their possible roles in focal

4

demyelination.135-137 It has been shown that LPC increases the oligomer formation of Aβ1-42

5

which is a key step for neurotoxicity63 and regulates the amyloid peptide fibril formation134 as

6

suggested with in vivo studies which might be associated in plaque formation in tgArcSwe

7

pathology. Further, LPC can induce focal demyelination and can disrupt the myelin lipids

8

nonspecifically.135 Therefore, pathological accumulations of LPC species on amyloid plaques

9

of tgArcSwe might be associated with cascaded reactions for focal amyloid plaque-

10

associated demyelination89 which is in line with plaque associated sulfatide depletion on the

11

very same amyloid plaques.(Figure 1) On the other hand, LPA is generated through several

12

enzymatic pathways such as their generation from membrane phospholipids through the

13

action of phospholipases138 and/or through the enzymatic action of autotaxin (ATX) which

14

metabolizes LPC to LPA.139 Further, LPA signals through cell-surface G protein-coupled

15

receptors (GPCRs), termed LPA receptors (LPARs) which are expressed in most cell types

16

including neurons, astrocytes, oligodendrocytes and schwann cells within CNS and have

17

been functionally linked to diverse physiological and pathophysiological processes within

18

nervous system140 such as myelin formation from CNS myelinating glia, oligodendrocytes.141

19

Indeed, increased ATX expression found in the frontal cortex of Alzheimer-type dementia

20

patients which suggests that ATX gene and cell signaling of LPA through LPAR signaling

21

might contribute to the disease pathology.142 Therefore, it might be inferred that damaged

22

nerve or glial cells can induce excessive production of LPA via ATX-mediated conversion of

23

LPC and/or hyperactivity of phospholipases which can trigger cascaded activations of

24

various LPARs in different cell types resulting in progressive demyelination. Furthermore,

25

LPA was found to induce increased Aβ production via upregulating BACE1 expression143

26

which can be their another possible role for amyloid plaque formation and progression.

27



Page 26 of 38

1

CONCLUSIONS

2

In summary, we have applied a multimodal MALDI-IMS method to reveal high throughput

3

molecular information on amyloid plaques at high spatial resolutions in 18 month-old

4

transgenic (tgArcSwe) AD mice brains. Inspection of ion images and multivariate image

5

analysis revealed plaque-associated alterations of amyloid peptide isoforms and lipids at

6

high spatial resolutions. On the other hand, lipid fragmentation obtained directly on plaque

7

aggregates at higher laser pulse energies provided tandem MS information useful for

8

structural elucidation of several lipid species. Dual polarity MALDI-IMS analysis of lipids on

9

the same pixel points revealed high troughput lipid molecular information which can be

10

correlated to individual amyloid plaques at a spatial resolution of 10 µm but not with 30 µm,

11

indicating the power of high spatial resolution IMS to probe spatially-confined plaque

12

features. We were, for the first time, able to correlate amyloid plaque-associated alterations

13

of several sphingolipids, phospholipids, and lysophospholipids to individual plaque features

14

and certain individual amyloid peptide isoforms on the same tissue section. Therefore, this

15

enabled us to comprehensively discuss the insights into the possible pathophysiological

16

pathways for amyloid plaque-associated alterations of lipids. Spatial alterations of several

17

lipid species and Aβ peptides reflected clues about the potential mechanisms of amyloid

18

plaque-associated

19

autophagic/lysosomal dysfunction, inflammation, oxidative stress and cell death.

20

EXPERIMENTAL SECTION

21

METHODS

22

Chemicals and Reagents. All chemicals for matrix and solvent preparation were pro-

23

analysis grade and obtained from Sigma-Aldrich (St. Louis, MO) unless otherwise specified.

24

TissueTek optimal cutting temperature (OCT) compound was purchased from Sakura

25

Finetek (AJ Alphen aan den Rijn, the Netherlands). The ddH2O was obtained from a Milli-Q

26

purification system (Merck Millipore, Darmstadt, Germany).

pathological

events

including

focal


demyelination,

astrogliosis,

Page 27 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

Animals, Tissue Sampling and Sectioning. Transgenic mice (n = 3), 18 months of age,

2

with the Arctic (E693G) and Swedish (K670N, M671L) mutations (tgArcSwe) of the human

3

APP were reared ad libitum at the animal facility at Uppsala University under a 12h/12h-

4

light/dark cycle. The animals were anesthetized with isoﬂurane and killed by decapitation.

5

The brains were dissected quickly with 3 min postmortem delay and frozen on dry ice. All

6

animal procedures were approved by an ethical committee and performed in compliance with

7

national and local animal care and use guidelines (DNr #C17⁄ 14 at Uppsala University).

8

Frozen tissue sections (12 µm thick) were cut in a cryostat microtome (Leica CM 1520, Leica

9

Biosystems, Nussloch, Germany) at -18°C, and collected on special-coated, conducting

10

glass slides (indium tin oxide, Bruker Daltonics, Bremen, Germany) and stored at -80°C.

11 12

Sample Preparation and Matrix Application Strategies. Prior to matrix deposition by

13

sublimation, tissue sections were thawed in a desiccator for ~30 minutes under reduced

14

pressure (SpeedVac, Eppendorf, Hamburg, Germany). For teaching of the navigation points

15

in Flex Imaging v3.0 software, we used a glass cutter pen (Starwalker, Montblanc, Germany)

16

to leave sharp-edged cross signs on tissue-mounted indium tin oxide (ITO) glass. The

17

sections were scanned in a slide scanner (PathScan Enabler IV, Electron Microscopy

18

Sciences, Hatﬁeld, PA, USA) to visualize the specific brain regions. Matrix deposition for lipid

19

analysis was carried out using a vacuum sublimation apparatus (Sigma-Aldrich) comprising

20

an inner flat top and an outer bottom attached to each other by an O-ring-sealed flange. The

21

chamber was connected to a rough pump attached to a digital vacuum gauge controller and

22

placed in a heated sand bath (SiO2, 50−70 mesh particle size, Sigma-Aldrich) on a hot plate

23

(C-MAG HP 4, IKA Werke GmbH & Co. KG, Staufen, Germany) connected to an electronic

24

contact thermometer (ETS-D5, IKA Werke GmbH & Co. KG, Staufen, Germany). Sublimation

25

was performed by the following steps. First, vacuum-dried (~20 minutes) ITO-glass slides

26

with the thaw-mounted mouse brain tissues were attached to the flat top of the chamber

27

using double-sided heat conductive copper tape (12mmx16.5m copper, 3M, United States).

28

Then, ~300 mg of 1,5-diaminonapthalene (1,5-DAN) matrix powder was spread evenly on



1

the outer bottom of the sublimation chamber, which was then attached to the top using the

2

O-ring seal. A vacuum of 0.8 mbar was provided by a membrane pump and the system was

3

allowed to equilibrate under vacuum for a period of ~20 minutes. The cooler was filled with

4

ice slush (≥0 °C) for condensation of the matrix on the sample slides. The process was

5

halted by removing the sublimation apparatus from the sand bath and closing the vacuum

6

valve. The sublimation apparatus was allowed to slowly re-equilibrate to atmospheric

7

pressure before removal of the sample plate. The amount of deposited matrix on ITO glass

8

was determined using a high-precision scale (AX224, Sartorius, Göttingen, Germany)

9

weighing the amount before and after the sublimation experiments. The sublimation protocol

10

was optimized with respect to temperature, deposition time and total amount of deposited

11

matrix to obtain the best detection efficiency for lipids on mouse brain tissue. With this setup,

12

the optimum matrix layer was found to be 120µg/cm2 to give the best lipid signals. We used

13

optimized sublimation conditions: 20 minutes at a temperature of 130°C under a stable

14

vacuum of 0.8 mbar. Prior to protein analysis, tissue sections were washed 2x 100% ethanol

15

(EtOH) (30s) to get rid of the remaining 1,5-DAN matrix molecules on the tissue surface.

16

Lipids and salts on the tissue surface were washed away in sequential washes of 70% EtOH

17

(30s), 100% EtOH (30s), Carnoy’s fluid (6:3:1 EtOH:chloroform:acetic acid) (2min), 100%

18

EtOH (30s), H2O with 0.2% TFA (30s), and 100% EtOH (30s).40 For teaching of the

19

navigation points we used the same sharped-edged cross points created with the glass

20

cutter pen prior to lipid analysis to minimize the repositioning error between lipid and protein

21

imaging analysis. For protein MALDI-IMS, 2,5-dihydroxyacetophenone (2,5-DHA) matrix

22

compound was applied using a TM Sprayer (HTX Technologies, Carrboro, NC, USA)

23

combined with a HPLC pump (Dionex P-580, Sunnyvale, CA, USA). Before every spraying

24

experiment, the corresponding pump valve was cleaned from big bubbles using a syringe

25

and ~10 minutes purge was applied to get rid of the little bubbles. Then the pump was kept

26

running at 100µl/min using a 70% ACN pushing solvent with isocratic pressure for 3 hours.

27

15mg/ml 2,5-DHA in 60% ACN/1%CH3COOH/1%TFA solution was sprayed over the tissue

28

sections using instrumental parameters including nitrogen flow of 10 psi, spray temperature


Page 28 of 38

Page 29 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

of 75°C, nozzle height of 40mm, 8 passes with offsets and rotations, a spray velocity of 1300

2

mm/min. These parameters were optimized to avoid analyte delocalization on the tissue

3

surface, while keeping the optimum incorporation of matrix and analyte molecules for an

4

efficient desorption/ionization.

5

MALDI-IMS Analysis. Imaging MS analysis of tissue sections was performed on a MALDI-

6

TOF/TOF UltrafleXtreme mass spectrometer equipped with SmartBeam II Nd:YAG/355 nm

7

laser operating at 1 kHz in TOF/TOF mode providing a laser spot diameter down to ∼10 µm

8

for the “minimum” focus setting (Bruker Daltonics).36 Measuring the absolute laser pulse

9

energy of a smartbeam II laser is challenging in UltrafleXtreme systems, as such a

10

measurement would require a reference measurement inside the instrument, which would

11

require modification of the instrument hardware. Moreover, SmartBeam lasers do not have a

12

flat-top shaped beam energy profile, but a rather highly structured beam energy profile,

13

which makes it a hard task to measure the exact value of laser fluence at a flat target

14

surface.144 Therefore, rather than measuring the laser pulse energy, we provide detailed

15

information about the laser pulse energy settings as previously indicated for lipid analysis in

16

our previous communication;36 global laser attenuator setting was kept stable at 10%

17

throughout all the experiments and the laser focus was set “minimum’’ for 10µm and “small’’

18

for 30µm spatial resolution analysis.

19

Attenuator offset and attenuator range settings were 40% and 10% for all the analysis. As

20

SmartBeam II lasers lose some energy output over the lifetime, it is necessary to specify the

21

laser shot count of the instrument unit used for this experimentation, which was about

22

2507888k (in ~13 months).

23

For dual polarity lipid analysis, profiling and imaging data acquisitions were performed in

24

reflective ion mode under optimized delayed extraction conditions in a mass range of 300-

25

3000 Da. First, IMS data acquired with 5 laser shots/per pixel point with a source

26

accelerating voltage of 20kV in negative polarity which was followed by 50 laser shots/pixel p



1

oint in positive polarity with a source accelerating voltage of 25kV. The detector gain value

2

was kept stable at 2626 V for both ionization modes. A mass resolution of M/∆M 20 000 was

3

achieved in the mass window of lipids (i.e., 650−1000 Da). External calibration was carried

4

out using peptide calibration standard I (Bruker Daltonics). Image data were reconstructed

5

and visualized using Flex Imaging v3.0 (Bruker Daltonics).

6

For peptide/protein analysis on the same imaging region of the same tissue section, global

7

laser attenuator was increased to 35% while keeping attenuator offset and range settings the

8

same. Profiling and imaging data acquisitions were performed in linear positive ion mode in a

9

mass range of 2000-20000 Da. Approximately 50 laser shots/raster spot were acquired with

10

1 kHz repetition rate.

11

Identification of amyloid β plaque-associated lipids was achieved by examining MS/MS

12

spectra obtained in LIFT-TOF/TOF mode as previously described.35,37 MALDI-LIFT (MS/MS)

13

spectra were collected directly from plaque deposits in LIFT mode until a satisfactory

14

convolution of fragment spectra was achieved. Lipid classifications were determined by

15

comparing mass accuracy data and fragment spectra with the LIPID MAPS database (Nature

16

Lipidomics Gateway, www.lipidmaps.org) and previous results obtained on tgArcSwe brain

17

tissue sections for the lipid species have already been reported.35,37 (Supporting Information

18

Figure S-4) Identification of amyloid peptides were determined comparing the mass accuracy

19

data with the previous study performed on the plaques of tgArcSwe mouse brain.33

20 21

Data Analysis. All MALDI-IMS data were calibrated externally using batch processing

22

function in Flex Analysis (v 3.0, Bruker Daltonics). Image analysis of the IMS data was

23

performed in SciLS (v2014, Bruker Daltonics). MALDI-IMS data were investigated with

24

principal component analysis (PCA) and hierarchical clustering analysis (HCA) based on

25

spatial segmentation (bisecting k-means) to probe the plaque-associated alterations of lipid

26

and peptide species.

27 28

ASSOCIATED CONTENT


Page 30 of 38

Page 31 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1 2

Supporting Information. Supplemental figures for multivariate image analysis and MS/MS

3

data for lipids.

4

AUTHOR INFORMATION

5 6

*Corresponding Author

7

Ibrahim Kaya

8

Department of Chemistry and Molecular Biology, University of Gothenburg, Analytical

9

Chemistry, Plan 4, Kemivägen 10, SE-412 96, Gothenburg, Sweden

10

[email protected];+46-73-716 86 04

11 12 13

ORCID: Jörg Hanrieder: 0000-0001-6059-198X 463

14

AUTHOR CONTRIBUTIONS I.K and J.H conceived and designed the study. I.K. designed

15

and performed the experiments. I.K and J.H. analyzed the data. I.K. interpreted the data and

16

wrote the major part of the manuscript. All the authors revised the manuscript and

17

contributed.

18 19

FUNDING The Swedish Research Council VR (#2014-6447, J.H.; #2012- 1593, S.S.; #2013-

20

2546, H.Z.; #2013-14002, K.B.), the European Research Council (#681712, H.Z.), Marie

21

Sklodowska Curie Actions (Cofund, Project INCA 600398, J.H.), the Royal Society of Arts

22

and Sciences in Gothenburg (KVVS, J.H.), Alzheimerfonden (J.H., K.B.), Åke Wiberg

23

Stiftelse (J.H.), Demensfonden (J.H.), Hjärnfonden (K.B.), Jeanssons Stiftelsen (J.H.), Ahlen

24

Stiftelsen (J.H.), Stiftelsen Gamla Tjänarinnor (J.H., K.B.), Stohnes Stiftelse (J.H.), and

25

Stiftelsen Wilhelm och Martina Lundgrens Vetenskapsfond (J.H.) are acknowledged for

26

financial support.

27 28

NOTES Authors declare no competing financial interest.

29



Page 32 of 38

1

ACKNOWLEDGEMENTS MALDI-IMS experiments were performed at the go:IMS imaging

2

MS infrastructure at the University of Gothenburg/Chalmers University of Technology

3

(www.go-ims.gu.se) headed by Prof. Andrew Ewing. We thank Dr. Stina Syvänen and Dr.

4

Dag Sehlin at Uppsala University for performing animal surgery and tissue preparations of

5

the tgArcSwe mice brain samples.

6 7

LIST OF ABBREVIATIONS ACN, acetonitrile; AD, Alzheimer’s disease; AGPS, alkyl-

8

dihydroxyacetone phosphate synthase; AICD, Aβ precursor protein intracellular domain;

9

ATX, autotaxin; AV, autophagic vacuoles; Aβ, amyloid beta; APP, Aβ precursor protein;

10

APOE, apolipoprotein E; BACE1, β-secretase 1; Cer, ceramides; CERK, ceramide kinase;

11

CH3COOH, acetic acid; CNS, central nervous system; DAN, diaminonapthalene; DHA,

12

dihydroxyacetophenone; EtOH, ethanol; ESI-MS, electrospray ionization mass spectrometry;

13

FA, fatty acid; FAD, familial AD; GM1, monosialotetrahexosylganglioside; GPCR, G protein-

14

coupled receptors; GSL, glycosphingolipid; HCA, hierarchical clustering analysis; h-FTAA,

15

heptamer-formyl thiophene acetic acid; HPLC, high-performance liquid chromatography;

16

IHC, immunohistochemistry; IMS, imaging mass spectrometry; ITO, indium tin-oxide; kV,

17

kilovoltage; LCO, luminescent conjugated oligothiophene; LPA, lysophosphatidic acid; LPAR,

18

lysophosphatidic

19

lysophosphatidylethanolamine; LPI, lysophosphatidylinositols; MALDI, matrix-assisted laser

20

desorption/ionization; MVA, multivariate analysis; NEP, neprilysin; NFT, neurofibrillary tangle;

21

OCT, optimal cutting temperature; PCA, principle component analysis; PA, phosphatidic

22

acid;

23

dimethylphosphatidylethanolamine; PI, phosphatidylinositols; PI3P, phosphatidylinositol-3-

24

phosphate;

25

PtdIns(4,5)P2, phosphatidylinositol-4,5-bisphosphate; PUFA, polyunsaturated fatty acid; RN,

26

reflector negative mode; ROI, region of interest; ROS, reactive oxygen species; RP, reflector

27

positive mode; SAD, sporadic AD; SiO2, silicon dioxide; SMases, sphingomyelinases; SM,

PC,

acid

receptor;

phosphatidylcholine;

PL,

phospholipase;

PG,

LPC,

PE,

lysophosphatidylcholine;

phosphatidylethanolamine;

phosphatidylglycerol;


PS,

LPE,

PE-NMe2,

phosphatidylserine;

Page 33 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

sphingomyelin; SPT, serine palmitoyl-CoA; ST, sulfatides; TFA, trifluoroacetic acid; TOF,

2

time-of-flight.

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

REFERENCES (1) Grundke-Iqbal, I.; Iqbal, K.; Tung, Y.-C.; Quinlan, M.; Wisniewski, H. M.; Binder, L. I. Proceedings of the National Academy of Sciences 1986, 83, 4913-4917. (2) Masters, C. L.; Simms, G.; Weinman, N. A.; Multhaup, G.; McDonald, B. L.; Beyreuther, K. Proceedings of the National Academy of Sciences 1985, 82, 4245-4249. (3) Foley, P. Biochimica et Biophysica Acta (BBA)-Molecular and Cell Biology of Lipids 2010, 1801, 750753. (4) Alzheimer, A. Allgemeine Zeitschrife Psychiatrie 1907, 64, 146-148. (5) Hardy, J. A.; Higgins, G. A. Science 1992, 256, 184. (6) Hardy, J.; Selkoe, D. J. science 2002, 297, 353-356. (7) Sakono, M.; Zako, T. The FEBS journal 2010, 277, 1348-1358. (8) Haass, C.; Selkoe, D. J. Nature reviews Molecular cell biology 2007, 8, 101-112. (9) Di Paolo, G.; Kim, T.-W. Nature Reviews Neuroscience 2011, 12, 284-296. (10) Simons, M.; Keller, P.; De Strooper, B.; Beyreuther, K.; Dotti, C. G.; Simons, K. Proceedings of the National Academy of Sciences 1998, 95, 6460-6464. (11) Cheng, H.; Vetrivel, K. S.; Gong, P.; Meckler, X.; Parent, A.; Thinakaran, G. Nature Clinical Practice Neurology 2007, 3, 374-382. (12) Sanchez-Mejia, R. O.; Newman, J. W.; Toh, S.; Yu, G.-Q.; Zhou, Y.; Halabisky, B.; Cissé, M.; Scearce-Levie, K.; Cheng, I. H.; Gan, L. Nature neuroscience 2008, 11, 1311-1318. (13) Berman, D. E.; Dall'Armi, C.; Voronov, S. V.; McIntire, L. B. J.; Zhang, H.; Moore, A. Z.; Staniszewski, A.; Arancio, O.; Kim, T.-W.; Di Paolo, G. Nature neuroscience 2008, 11, 547-554. (14) Oliveira, T. G.; Chan, R. B.; Tian, H.; Laredo, M.; Shui, G.; Staniszewski, A.; Zhang, H.; Wang, L.; Kim, T.-W.; Duff, K. E. Journal of Neuroscience 2010, 30, 16419-16428. (15) Zha, Q.; Ruan, Y.; Hartmann, T.; Beyreuther, K.; Zhang, D. Molecular psychiatry 2004, 9, 946-952. (16) Yanagisawa, K.; Odaka, A.; Suzuki, N.; Ihara, Y. Nature medicine 1995, 1, 1062-1066. (17) Corder, E. H.; Saunders, A. M.; Strittmatter, W. J.; Schmechel, D. E.; Gaskell, P. C.; Small, G.; Roses, A. D.; Haines, J.; Pericak-Vance, M. A. Science 1993, 261, 921-923. (18) Bu, G. Nature Reviews Neuroscience 2009, 10, 333-344. (19) Selkoe, D. J. Neuron 1991, 6, 487-498. (20) Grimm, M. O.; Mett, J.; Grimm, H. S.; Hartmann, T. Frontiers in Molecular Neuroscience 2017, 10. (21) Panchal, M.; Loeper, J.; Cossec, J.-C.; Perruchini, C.; Lazar, A.; Pompon, D.; Duyckaerts, C. Journal of lipid research 2010, 51, 598-605. (22) Han, X.; Holtzman, D. M.; McKeel, D. W. Journal of neurochemistry 2001, 77, 1168-1180. (23) Han, X.; M Holtzman, D.; W McKeel, D.; Kelley, J.; Morris, J. C. Journal of neurochemistry 2002, 82, 809-818. (24) Portelius, E.; Bogdanovic, N.; Gustavsson, M. K.; Volkmann, I.; Brinkmalm, G.; Zetterberg, H.; Winblad, B.; Blennow, K. Acta neuropathologica 2010, 120, 185-193. (25) Gurel, B.; Cansev, M.; Sevinc, C.; Kelestemur, S.; Ocalan, B.; Cakir, A.; Aydin, S.; Kahveci, N.; Ozansoy, M.; Taskapilioglu, O. Journal of Alzheimer's Disease 2018, 61, 1399-1410. (26) Sullivan, P.; Han, B.; Liu, F.; Mace, B.; Ervin, J.; Wu, S.; Koger, D.; Paul, S.; Bales, K. Neurobiology of aging 2011, 32, 791-801. (27) McDonnell, L. A.; Heeren, R. Mass spectrometry reviews 2007, 26, 606-643. (28) Stoeckli, M.; Staab, D.; Staufenbiel, M.; Wiederhold, K.-H.; Signor, L. Analytical biochemistry 2002, 311, 33-39. (29) Kelley, A. R.; Perry, G.; Castellani, R. J.; Bach, S. B. ACS chemical neuroscience 2016, 7, 261-268.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

(30) Solé-Domènech, S.; Sjövall, P.; Vukojević, V.; Fernando, R.; Codita, A.; Salve, S.; Bogdanović, N.; Mohammed, A. H.; Hammarström, P.; Nilsson, K. P. R. Acta neuropathologica 2013, 125, 145-157. (31) Carlred, L.; Gunnarsson, A.; Solé-Domènech, S.; Johansson, B. r.; Vukojević, V.; Terenius, L.; Codita, A.; Winblad, B.; Schalling, M.; Höök, F. Journal of the American Chemical Society 2014, 136, 9973-9981. (32) Carlred, L.; Vukojević, V.; Johansson, B.; Schalling, M.; Höök, F.; Sjövall, P. Biointerphases 2016, 11, 02A312. (33) Carlred, L.; Michno, W.; Kaya, I.; Sjövall, P.; Syvänen, S.; Hanrieder, J. Journal of neurochemistry 2016, 138, 469-478. (34) Kakuda, N.; Miyasaka, T.; Iwasaki, N.; Nirasawa, T.; Wada-Kakuda, S.; Takahashi-Fujigasaki, J.; Murayama, S.; Ihara, Y.; Ikegawa, M. Acta neuropathologica communications 2017, 5, 73. (35) Kaya, I.; Brinet, D.; Michno, W.; Syvänen, S.; Sehlin, D.; Zetterberg, H.; Blennow, K.; Hanrieder, J. r. ACS chemical neuroscience 2017, 8, 347-355. (36) Kaya, I.; Michno, W.; Brinet, D.; Iacone, Y.; Zanni, G.; Blennow, K.; Zetterberg, H.; Hanrieder, J. r. Analytical Chemistry 2017, 89, 4685-4694. (37) Kaya, I.; Brinet, D.; Michno, W.; Başkurt, M.; Zetterberg, H.; Blenow, K.; Hanrieder, J. r. ACS chemical neuroscience 2017, 8, 2778-2790. (38) Selkoe, D. J. Journal of Biological Chemistry 1996, 271, 18295-18298. (39) Lord, A.; Kalimo, H.; Eckman, C.; Zhang, X.-Q.; Lannfelt, L.; Nilsson, L. N. Neurobiology of aging 2006, 27, 67-77. (40) Yang, J.; Caprioli, R. M. Analytical chemistry 2011, 83, 5728-5734. (41) Thomas, A. l.; Charbonneau, J. L.; Fournaise, E.; Chaurand, P. Analytical chemistry 2012, 84, 2048-2054. (42) Spraggins, J. M.; Rizzo, D. G.; Moore, J. L.; Rose, K. L.; Hammer, N. D.; Skaar, E. P.; Caprioli, R. M. Journal of the American Society for Mass Spectrometry 2015, 26, 974-985. (43) Spraggins, J. M.; Rizzo, D. G.; Moore, J. L.; Noto, M. J.; Skaar, E. P.; Caprioli, R. M. Proteomics 2016, 16, 1678-1689. (44) Molin, L.; Seraglia, R.; Dani, F. R.; Moneti, G.; Traldi, P. Rapid Communications in Mass Spectrometry 2011, 25, 3091-3096. (45) Jackson, S. N.; Wang, H.-Y. J.; Woods, A. S.; Ugarov, M.; Egan, T.; Schultz, J. A. Journal of the American Society for Mass Spectrometry 2005, 16, 133-138. (46) Klingstedt, T.; Blechschmidt, C.; Nogalska, A.; Prokop, S.; Häggqvist, B.; Danielsson, O.; Engel, W. K.; Askanas, V.; Heppner, F. L.; Nilsson, K. P. R. ChemBioChem 2013, 14, 607-616. (47) Schaeffer, E. L.; Figueiro, M.; Gattaz, W. F. Clinics 2011, 66, 45-54. (48) Kuo, Y.-M.; Kokjohn, T. A.; Beach, T. G.; Sue, L. I.; Brune, D.; Lopez, J. C.; Kalback, W. M.; Abramowski, D.; Sturchler-Pierrat, C.; Staufenbiel, M. Journal of Biological Chemistry 2001, 276, 12991-12998. (49) Kelley, A. R.; Perry, G.; Bach, S. B. ACS chemical neuroscience 2018. (50) Güntert, A.; Döbeli, H.; Bohrmann, B. Neuroscience 2006, 143, 461-475. (51) Condello, C.; Lemmin, T.; Stöhr, J.; Nick, M.; Wu, Y.; Maxwell, A. M.; Watts, J. C.; Caro, C. D.; Oehler, A.; Keene, C. D. Proceedings of the National Academy of Sciences 2018, 201714966. (52) Zavalin, A.; Yang, J.; Hayden, K.; Vestal, M.; Caprioli, R. M. Analytical and bioanalytical chemistry 2015, 407, 2337-2342. (53) Kompauer, M.; Heiles, S.; Spengler, B. nAture methods 2017, 14, 90. (54) Kojro, E.; Gimpl, G.; Lammich, S.; März, W.; Fahrenholz, F. Proceedings of the National Academy of Sciences 2001, 98, 5815-5820. (55) Demuro, A.; Mina, E.; Kayed, R.; Milton, S. C.; Parker, I.; Glabe, C. G. Journal of Biological Chemistry 2005, 280, 17294-17300. (56) Grimm, M. O.; Mett, J.; Stahlmann, C. P.; Grösgen, S.; Haupenthal, V. J.; Blümel, T.; Hundsdörfer, B.; Zimmer, V. C.; Mylonas, N. T.; Tanila, H. Frontiers in aging neuroscience 2015, 7.


Page 34 of 38

Page 35 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

(57) Grimm, M. O.; Grösgen, S.; Rothhaar, T. L.; Burg, V. K.; Hundsdörfer, B.; Haupenthal, V. J.; Friess, P.; Müller, U.; Fassbender, K.; Riemenschneider, M. International Journal of Alzheimer’s Disease 2011, 2011. (58) Grimm, M. O.; Kuchenbecker, J.; Rothhaar, T. L.; Grösgen, S.; Hundsdörfer, B.; Burg, V. K.; Friess, P.; Müller, U.; Grimm, H. S.; Riemenschneider, M. Journal of neurochemistry 2011, 116, 916-925. (59) Grimm, M. O.; Hundsdörfer, B.; Grösgen, S.; Mett, J.; Zimmer, V. C.; Stahlmann, C. P.; Haupenthal, V. J.; Rothhaar, T. L.; Lehmann, J.; Pätzold, A. Cellular Physiology and Biochemistry 2014, 34, 92-110. (60) Grimm, M. O.; Grimm, H. S.; Pätzold, A. J.; Zinser, E. G.; Halonen, R.; Duering, M.; Tschäpe, J.-A.; De Strooper, B.; Müller, U.; Shen, J. Nature cell biology 2005, 7, 1118-1123. (61) Tamboli, I. Y.; Prager, K.; Barth, E.; Heneka, M.; Sandhoff, K.; Walter, J. Journal of Biological Chemistry 2005, 280, 28110-28117. (62) Onodera, T.; Futai, E.; Kan, E.; Abe, N.; Uchida, T.; Kamio, Y.; Kaneko, J. The Journal of Biochemistry 2014, 157, 301-309. (63) Sheikh, A. M.; Michikawa, M.; Kim, S.; Nagai, A. Neuroscience 2015, 292, 159-169. (64) McLaurin, J.; Franklin, T.; Chakrabartty, A.; Fraser, P. Journal of molecular biology 1998, 278, 183194. (65) Han, X.; Cheng, H.; Fryer, J. D.; Fagan, A. M.; Holtzman, D. M. Journal of Biological Chemistry 2003, 278, 8043-8051. (66) Liu, C.-C.; Kanekiyo, T.; Xu, H.; Bu, G. Nature Reviews Neurology 2013, 9, 106. (67) Hannun, Y. A.; Obeid, L. M. Nature reviews Molecular cell biology 2008, 9, 139-150. (68) Puglielli, L.; Ellis, B. C.; Saunders, A. J.; Kovacs, D. M. Journal of Biological Chemistry 2003, 278, 19777-19783. (69) He, X.; Huang, Y.; Li, B.; Gong, C.-X.; Schuchman, E. H. Neurobiology of aging 2010, 31, 398-408. (70) Haughey, N. J.; Bandaru, V. V.; Bae, M.; Mattson, M. P. Biochimica et Biophysica Acta (BBA)Molecular and Cell Biology of Lipids 2010, 1801, 878-886. (71) de San Román, E. G.; Manuel, I.; Giralt, M.; Ferrer, I.; Rodríguez-Puertas, R. Biochimica et Biophysica Acta (BBA)-Biomembranes 2017, 1859, 1604-1614. (72) Isaac, G.; Pernber, Z.; Gieselmann, V.; Hansson, E.; Bergquist, J.; Månsson, J. E. The FEBS journal 2006, 273, 1782-1790. (73) Hirahara, Y.; Wakabayashi, T.; Mori, T.; Koike, T.; Yao, I.; Tsuda, M.; Honke, K.; Gotoh, H.; Ono, K.; Yamada, H. Journal of neurochemistry 2017, 140, 435-450. (74) Marcus, J.; Honigbaum, S.; Shroff, S.; Honke, K.; Rosenbluth, J.; Dupree, J. Glia 2006, 53, 372-381. (75) Cheng, H.; Zhou, Y.; Holtzman, D. M.; Han, X. Neurobiology of aging 2010, 31, 1188-1196. (76) Han, X. Journal of neurochemistry 2007, 103, 171-179. (77) Cutler, R. G.; Kelly, J.; Storie, K.; Pedersen, W. A.; Tammara, A.; Hatanpaa, K.; Troncoso, J. C.; Mattson, M. P. Proceedings of the National Academy of Sciences 2004, 101, 2070-2075. (78) Jana, A.; Pahan, K. Journal of Biological Chemistry 2004, 279, 51451-51459. (79) Lee, J.-T.; Xu, J.; Lee, J.-M.; Ku, G.; Han, X.; Yang, D.-I.; Chen, S.; Hsu, C. Y. The Journal of cell biology 2004, 164, 123-131. (80) Katsel, P.; Li, C.; Haroutunian, V. Neurochemical research 2007, 32, 845-856. (81) Back, S. A.; Gan, X.; Li, Y.; Rosenberg, P. A.; Volpe, J. J. Journal of Neuroscience 1998, 18, 62416253. (82) Al-Mashhadi, S.; Simpson, J. E.; Heath, P. R.; Dickman, M.; Forster, G.; Matthews, F. E.; Brayne, C.; Ince, P. G.; Wharton, S. B. Brain Pathology 2015, 25, 565-574. (83) Xu, J.; Chen, S.; Ahmed, S. H.; Chen, H.; Ku, G.; Goldberg, M. P.; Hsu, C. Y. Journal of Neuroscience 2001, 21, RC118-RC118. (84) Butterfield, D. A.; Castegna, A.; Lauderback, C. M.; Drake, J. Neurobiology of aging 2002, 23, 655664. (85) Obeid, L. M.; Linardic, C. M.; Karolak, L. A.; Hannun, Y. A. Science 1993, 259, 1769-1771. (86) Toman, R. E.; Movsesyan, V.; Murthy, S. K.; Milstien, S.; Spiegel, S.; Faden, A. I. Journal of neuroscience research 2002, 68, 323-330.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51

(87) Gómez-Muñoz, A. Biochimica et Biophysica Acta (BBA)-Biomembranes 2006, 1758, 2049-2056. (88) Chalfant, C. E.; Spiegel, S. Journal of cell science 2005, 118, 4605-4612. (89) Mitew, S.; Kirkcaldie, M. T.; Halliday, G. M.; Shepherd, C. E.; Vickers, J. C.; Dickson, T. C. Acta neuropathologica 2010, 119, 567-577. (90) C Schmued, L.; Raymick, J.; G Paule, M.; Dumas, M.; Sarkar, S. Current Alzheimer Research 2013, 10, 30-37. (91) Benitez, A.; Fieremans, E.; Jensen, J. H.; Falangola, M. F.; Tabesh, A.; Ferris, S. H.; Helpern, J. A. NeuroImage: Clinical 2014, 4, 64-71. (92) Papuć, E.; Rejdak, K. J Alzheimers Dis Parkinsonism 2017, 7, 2161-0460.1000321. (93) Svennerholm, L.; Gottfries, C. G. Journal of neurochemistry 1994, 62, 1039-1047. (94) Crino, P. B.; Ullman, M. D.; Vogt, B. A.; Bird, E. D.; Volicer, L. Archives of neurology 1989, 46, 398401. (95) Barrier, L.; Ingrand, S.; Damjanac, M.; Bilan, A. R.; Hugon, J.; Page, G. Neurobiology of aging 2007, 28, 1863-1872. (96) Kracun, I.; Kalanj, S.; Talan-Hranilovic, J.; Cosovic, C. Neurochemistry international 1992, 20, 433438. (97) Kreutz, F.; dos Santos Petry, F.; Camassola, M.; Schein, V.; Guma, F. C.; Nardi, N. B.; Trindade, V. M. T. Gene 2013, 527, 109-114. (98) Dufresne, M.; Guneysu, D.; Patterson, N. H.; Marcinkiewicz, M. M.; Regina, A.; Demeule, M.; Chaurand, P. Analytical and Bioanalytical Chemistry 2016, 1-9. (99) Maragakis, N. J.; Rothstein, J. D. Nature Reviews Neurology 2006, 2, 679. (100) Olsen, M.; Aguilar, X.; Sehlin, D.; Fang, X. T.; Antoni, G.; Erlandsson, A.; Syvänen, S. Molecular Imaging and Biology 2018, 1-10. (101) Nixon, R. A.; Cataldo, A. M. Journal of Alzheimer's Disease 2006, 9, 277-289. (102) Ginsberg, S. D.; Alldred, M. J.; Counts, S. E.; Cataldo, A. M.; Neve, R. L.; Jiang, Y.; Wuu, J.; Chao, M. V.; Mufson, E. J.; Nixon, R. A. Biological psychiatry 2010, 68, 885-893. (103) Orr, M. E.; Oddo, S. Alzheimer's research & therapy 2013, 5, 53. (104) Nixon, R. A.; Wegiel, J.; Kumar, A.; Yu, W. H.; Peterhoff, C.; Cataldo, A.; Cuervo, A. M. Journal of Neuropathology & Experimental Neurology 2005, 64, 113-122. (105) Son, J. H.; Shim, J. H.; Kim, K.-H.; Ha, J.-Y.; Han, J. Y. Experimental & molecular medicine 2012, 44, 89. (106) Boland, B.; Kumar, A.; Lee, S.; Platt, F. M.; Wegiel, J.; Yu, W. H.; Nixon, R. A. Journal of Neuroscience 2008, 28, 6926-6937. (107) Nixon, R. A. Journal of cell science 2007, 120, 4081-4091. (108) Farooqui, A. A.; Horrocks, L. A.; Farooqui, T. Chemistry and physics of lipids 2000, 106, 1-29. (109) Farooqui, A. A.; Horrocks, L. A.; Farooqui, T. Journal of Molecular Neuroscience 2000, 14, 123135. (110) Kern, R.; Joseleau-Petit, D.; Chattopadhyay, M. K.; Richarme, G. Biochemical and biophysical research communications 2001, 289, 1268-1274. (111) Roßner, S. International journal of developmental neuroscience 2004, 22, 467-474. (112) Cai, D.; Netzer, W. J.; Zhong, M.; Lin, Y.; Du, G.; Frohman, M.; Foster, D. A.; Sisodia, S. S.; Xu, H.; Gorelick, F. S. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 1941-1946. (113) Stokes, C. E.; Hawthorne, J. N. Journal of neurochemistry 1987, 48, 1018-1021. (114) Oliveira, T. G.; Di Paolo, G. Biochimica et Biophysica Acta (BBA)-Molecular and Cell Biology of Lipids 2010, 1801, 799-805. (115) Ross, B. M.; Moszczynska, A.; Erlich, J.; Kish, S. J. Journal of neurochemistry 1998, 70, 786-793. (116) Hong, J. H.; Kang, J. W.; Kim, D. K.; Baik, S. H.; Kim, K. H.; Shanta, S. R.; Jung, J. H.; Mook-Jung, I.; Kim, K. P. Journal of lipid research 2016, 57, 36-45. (117) Jolles, J.; Bothmer, J.; Markerink, M.; Ravid, R. Journal of neurochemistry 1992, 58, 2326-2329. (118) Ellis, S. R.; Cappell, J.; Potočnik, N. O.; Balluff, B.; Hamaide, J.; Van der Linden, A.; Heeren, R. M. Analyst 2016.


Page 36 of 38

Page 37 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

(119) Ginsberg, L.; Rafique, S.; Xuereb, J. H.; Rapoport, S. I.; Gershfeld, N. L. Brain research 1995, 698, 223-226. (120) Igarashi, M.; Ma, K.; Gao, F.; Kim, H.-W.; Rapoport, S. I.; Rao, J. S. Journal of Alzheimer's Disease 2011, 24, 507-517. (121) Sanchez-Mejia, R. O.; Newman, J. W.; Toh, S.; Yu, G.-Q.; Zhou, Y.; Halabisky, B.; Cissé, M.; Scearce-Levie, K.; Cheng, I. H.; Gan, L. Nature neuroscience 2008, 11, 1311. (122) Butterfield, D. A.; Drake, J.; Pocernich, C.; Castegna, A. Trends in molecular medicine 2001, 7, 548-554. (123) Butterfield, D. A.; Lauderback, C. M. Free Radical Biology and Medicine 2002, 32, 1050-1060. (124) Furman, R.; Murray, I. V.; Schall, H. E.; Liu, Q.; Ghiwot, Y.; Axelsen, P. H. ACS chemical neuroscience 2016, 7, 367-377. (125) De Vet, E.; Ijlst, L.; Oostheim, W.; Dekker, C.; Moser, H.; van Den Bosch, H.; Wanders, R. Journal of lipid research 1999, 40, 1998-2003. (126) Avdulov, N. A.; Chochina, S. V.; Igbavboa, U.; Warden, C. S.; Vassiliev, A. V.; Wood, W. G. Journal of neurochemistry 1997, 69, 1746-1752. (127) Chan, R. B.; Oliveira, T. G.; Cortes, E. P.; Honig, L. S.; Duff, K. E.; Small, S. A.; Wenk, M. R.; Shui, G.; Di Paolo, G. Journal of Biological Chemistry 2012, 287, 2678-2688. (128) Di Paolo, G.; De Camilli, P. Nature 2006, 443, 651. (129) Morel, E.; Chamoun, Z.; Lasiecka, Z. M.; Chan, R. B.; Williamson, R. L.; Vetanovetz, C.; Dall’Armi, C.; Simoes, S.; Du Jour, K. S. P.; McCabe, B. D. Nature communications 2013, 4, 2250. (130) Miranda, A. M.; Lasiecka, Z. M.; Xu, Y.; Neufeld, J.; Shahriar, S.; Simoes, S.; Chan, R. B.; Oliveira, T. G.; Small, S. A.; Di Paolo, G. Nature communications 2018, 9, 291. (131) Kudo, I.; Murakami, M. Prostaglandins & other lipid mediators 2002, 68, 3-58. (132) Sundaram, J. R.; Chan, E. S.; Poore, C. P.; Pareek, T. K.; Cheong, W. F.; Shui, G.; Tang, N.; Low, C.M.; Wenk, M. R.; Kesavapany, S. Journal of Neuroscience 2012, 32, 1020-1034. (133) Stephenson, D. T.; Lemere, C. A.; Selkoe, D. J.; Clemens, J. A. Neurobiology of disease 1996, 3, 51-63. (134) Sheikh, A. M.; Nagai, A. FEBS journal 2011, 278, 634-642. (135) Plemel, J. R.; Michaels, N. J.; Weishaupt, N.; Caprariello, A. V.; Keough, M. B.; Rogers, J. A.; Yukseloglu, A.; Lim, J.; Patel, V. V.; Rawji, K. S. Glia 2018, 66, 327-347. (136) Allt, G.; Ghabriel, M.; Sikri, K. Acta neuropathologica 1988, 75, 456-464. (137) HALL, S. M. Journal of cell science 1972, 10, 535-546. (138) Aoki, J.; Taira, A.; Takanezawa, Y.; Kishi, Y.; Hama, K.; Kishimoto, T.; Mizuno, K.; Saku, K.; Taguchi, R.; Arai, H. Journal of Biological Chemistry 2002, 277, 48737-48744. (139) Tanaka, M.; Okudaira, S.; Kishi, Y.; Ohkawa, R.; Iseki, S.; Ota, M.; Noji, S.; Yatomi, Y.; Aoki, J.; Arai, H. Journal of Biological Chemistry 2006, 281, 25822-25830. (140) Yung, Y. C.; Stoddard, N. C.; Mirendil, H.; Chun, J. Neuron 2015, 85, 669-682. (141) Weiner, J. A.; Hecht, J. H.; Chun, J. The Journal of comparative neurology 1998, 398, 587-598. (142) Umemura, K.; Yamashita, N.; Yu, X.; Arima, K.; Asada, T.; Makifuchi, T.; Murayama, S.; Saito, Y.; Kanamaru, K.; Goto, Y. Neuroscience letters 2006, 400, 97-100. (143) Shi, J.; Dong, Y.; Cui, M.-Z.; Xu, X. Biochimica et Biophysica Acta (BBA)-Molecular Basis of Disease 2013, 1832, 29-38. (144) Optimizing, U. Journal of Mass Spectrometry 2006, 41, 705-716.

44 45

For TOC figure only:



1


Page 38 of 38

Shedding Light on the Molecular Pathology of Amyloid Plaques in

Recommend Documents