Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Dec 28;169(12):e70335. doi: 10.1111/jnc.70335

Post‐Translational Modifications Distinguish Amyloid‐β Isoforms in Cerebral Amyloid Angiopathy and Alzheimer's Disease

Srinivas Koutarapu 1, Kaleigh F Roberts 1, Reid A Coyle 2,3, Jogender Mehla 4, Chihiro Sato 2,3, Gregory J Zipfel 4, Randall J Bateman 2,3, Katherine E Schwetye 1,✉, Soumya Mukherjee 2,3,✉
PMCID: PMC12745913  PMID: 41457650

ABSTRACT

Cerebral amyloid angiopathy (CAA) shares amyloid‐β (Aβ) deposition as a pathological hallmark with the extracellular plaques of Alzheimer's disease (AD). While both disease processes involve progressive, decades‐long deposition of fibrillar Aβ peptide, they differ in isoform composition. We hypothesized that post‐translational modifications (PTMs) on Aβ would also differ between CAA and parenchymal plaques. Using Lys‐N enzymatic digestion followed by quantitative mass spectrometry, we profiled Aβ isoforms and N‐terminus PTMs (aspartic acid isomerization and pyroglutamate formation) across CAA severity and compared them to parenchymal plaque Aβ in AD. Moderate to severe CAA were dominated by intact N‐terminus (Aβ1‐x ~ 95%) with minimal N‐truncated species (Aβ2‐x, Aβ3pGlu‐x, and Aβ4‐x), whereas parenchymal plaques displayed diverse N‐terminus truncations and PTMs. Increasing CAA severity correlated with a shift from longer, hydrophobic C‐terminal isoforms (Aβ41, Aβ42, and Aβ43) to shorter, less hydrophobic C‐terminal isoforms (Aβ37, Aβ38, Aβ39, and Aβ40). Importantly, moderate and severe CAA displayed minimal isomerization of Asp1 and Asp7 residues. These patterns suggest distinct Aβ aggregation mechanisms in CAA versus parenchymal plaques. We propose that the intact and unmodified N‐terminus found in CAA is due to its inclusion within the protofibril structure making them less disordered and inaccessible to post‐translational modifications, in contrast to plaque‐associated Aβ. These biochemical differences may reflect underlying structural distinctions in protofibril architectures, with potential implications for biomarker development for early CAA detection and therapeutic targeting of vascular versus parenchymal Aβ.

graphic file with name JNC-169-0-g003.jpg

Keywords: Alzheimer's disease, amyloid‐β, cerebral amyloid angiopathy, mass spectrometry, parenchymal plaques

Cerebral amyloid angiopathy (CAA) is characterized by progressive vascular deposition of fibrillar Aβ. Using Lys‐N digestion and quantitative mass spectrometry, we profiled Aβ isoforms and N‐terminal post‐translational modifications (PTMs) across CAA severities. Moderate to severe CAA predominantly contained intact Aβ1‐x (~95%) with minimal N‐terminal truncations, whereas parenchymal plaques exhibited diverse truncations and PTMs. We propose that the intact N‐terminus in CAA reflects its structural embedding within ordered protofibrils, rendering it less accessible to PTMs. These biochemical distinctions suggest differing fibril architectures and may guide biomarker and therapeutic strategies for vascular versus parenchymal Aβ pathology.

graphic file with name JNC-169-0-g004.jpg

Abbreviations

AD

Alzheimer's disease

APP

amyloid precursor protein

Arg

arginine

ARIA

amyloid‐related imaging abnormalities

Asp

aspartate

Aβ

amyloid‐β

CAA

cerebral amyloid angiopathy

EM

electron microscopy

FA

formic acid

Gly

glycine

IHC

immunohistochemistry

IP/MS

immunoprecipitation based mass spectrometry

Knight ADRC

Knight Alzheimer Disease Research Center

MALDI‐MS imaging

matrix‐assisted laser desorption ionization‐mass spectrometry imaging

pGlu

pyroglutamate

PTMs

post‐translational modifications

TEABC

triethylammonium bicarbonate buffer

1. Introduction

Cerebral amyloid angiopathy (CAA) shares amyloid‐β (Aβ) deposition as a pathological hallmark with the extracellular plaques of Alzheimer's disease (AD) and occurs in 85%–95% of AD patients (Masters et al. 1985; Goedert et al. 1989; Walker et al. 2024; Yamada 2000; Keage et al. 2009). Fibrillar Aβ is the main component of both parenchymal plaques of AD and vascular deposits of CAA (Masters et al. 1985; Roher, Lowenson, Clarke, Wolkow, et al. 1993). In CAA, Aβ deposition begins within the basement membrane and progresses into the tunica media of cerebral and leptomeningeal vessels. In advanced cases, extensive vascular Aβ accumulation may lead to fibrinoid necrosis and microaneurysm formation, ultimately compromising vessel integrity and increasing the risk of micro‐ and lobar hemorrhages (Attems et al. 2011). Depending on the drug and clinical trial design, CAA has been linked to increased amyloid‐related imaging abnormalities (ARIA) in 20%–40% of the patients receiving anti‐amyloid targeting therapies (Mintun et al. 2021; Van Dyck et al. 2023). Understanding the molecular signatures of CAA is crucial for developing effective therapies and identifying biomarkers to predict which AD patients are at higher risk of treatment‐related complications.

Aβ peptides that deposit as aggregates in the brain are usually 38–43 amino acids in length and result from the sequential cleavage of the amyloid precursor protein (APP) by β‐secretase and γ‐secretase enzymes (Vassar et al. 1999). N‐terminal and C‐terminal processing of Aβ results in the formation of various proteoforms with altered physicochemical properties, which can influence seeding and aggregation propensity (He and Barrow 1999) and contribute to the formation of pathological deposits (Cabrera et al. 2018). While Aβ1–40 and Aβ1–42 are the most common isoforms found in CAA and AD parenchymal plaques, respectively (Roher, Lowenson, Clarke, Woods, et al. 1993; Kakuda et al. 2017; Miller et al. 1993), numerous other variants have been identified in postmortem AD brain tissue (Gkanatsiou et al. 2019; Portelius et al. 2010; Reinert et al. 2016). Recent studies using advanced chemical imaging approaches, such as the combination of MALDI‐mass spectrometry imaging and immunohistochemistry, have revealed distinct differences in the composition of Aβ deposits in the cerebral vasculature and parenchymal regions of human brain (Kasri et al. 2024; Koutarapu et al. 2025).

Additionally, multiple post‐translational modifications (PTMs) of Aβ have been characterized, which further diversify the proteoforms of Aβ contributing to these pathologies (Lyons et al. 2016; Michno et al. 2019). The majority of Aβ PTMs extracted from the plaques are localized on the N‐terminus, including spontaneous nonenzymatic isomerization of the aspartic acid residues (Roher, Lowenson, Clarke, Wolkow, et al. 1993), pyroglutamate formation (Kuo et al. 1997; Michno et al. 2019; Saido et al. 1995), and arginine citrullination (Mukherjee, Perez, Dubois, et al. 2021; Mukherjee, Perez, Lago, et al. 2021).

Isomerization at Asp7 is abundant in both parenchymal plaque and amyloid deposited vasculature (Shin et al. 2003; Tomidokoro et al. 2010; Roher, Lowenson, Clarke, Woods, et al. 1993), while isomerization at Asp23 is mainly associated with heavy vascular amyloidosis associated with dementia and intracerebral hemorrhages (Fossati et al. 2013). Substantial isomerization at Asp1 and Asp7 has been reported in Aβ extracted from parenchymal plaques, with levels exceeding 90% in AD and approximately 50% in preclinical AD (Mukherjee et al. 2025; Mukherjee, Perez, Lago, et al. 2021; Schrempel et al. 2024). The extent of Asp1 and Asp7 isomerization is related to the age of the aggregates, suggesting that the parenchymal plaque Aβ accumulates these modifications over time as plaques mature (Roher, Lowenson, Clarke, Wolkow, et al. 1993; Lyons et al. 2016; Lambeth et al. 2019). However, the degree to which these Asp residues are modified in CAA remains unknown. Given that these PTMs have been shown to alter physicochemical properties, including toxicity (Crehan et al. 2020; Lambeth et al. 2019), characterization of how PTM patterns differ between vascular and parenchymal Aβ is critical to understanding disease pathogenesis and may help predict differential responses to anti‐amyloid therapeutics, including the risk for adverse side effects such as ARIA.

In this study, we utilized a clinically well‐characterized frozen human brain tissue cohort to investigate Aβ proteoform distributions across neuropathologically classified CAA severities: none, mild, moderate, and severe. We isolated cortical vessels using density gradient centrifugation and fractionated the vascular proteins into soluble and insoluble pools. We used immunoprecipitation coupled with liquid chromatography mass spectrometry (IP/MS) on both fractions. We characterized the N‐ and C‐terminal isoform heterogeneity across CAA severities and compared them to Aβ derived from insoluble parenchymal plaques from AD. In addition to identifying distinct Aβ isoform distributions in vascular versus parenchymal plaques, we also characterized PTMs including aspartic acid isomerization at Asp1 and Asp7, N‐terminal pyroglutamate formation at pGlu3, and arginine citrullination at Arg5. The results suggest that the origin of the N‐ and C‐terminal heterogeneity and PTMs in vascular versus parenchymal plaque Aβ may reflect differences in the protofibril structure within these distinct aggregation environments. Our results have implications for better therapeutic strategies for targeting CAA as well as diagnostic biomarker development.

2. Materials and Methods

2.1. Samples

Autopsy brain samples from AD patients with mild, moderate, or severe CAA (n = 5 each), AD patients without CAA (n = 5), and age‐matched cognitively healthy controls (n = 5) were obtained from the Charles F. and Joanne Knight Alzheimer Disease Research Center (Knight ADRC) at Washington University in St. Louis. AD patients without CAA and age‐matched cognitively healthy controls were grouped together as the CAA‐none cohort (Refer to Table 1 for demographics). In order to estimate N‐terminal (Asp1 and Asp7) isomerization in parenchymal plaques, we reanalyzed data from 11 ad patients from (Mukherjee, Perez, Lago, et al. 2021). For distributions of isoforms within parenchymal plaques, we used data derived from 2 ad patients from (Horie et al. 2020) where all isoforms of interest were measured. Institutional/ethical approval number: ADRC (201105102), ACS (201105305), and HASD (201105103). Detailed demographics are available in Table S2. No sample size calculation was performed a priori.

TABLE 1.

Demographics.

CAA severity Age (years) Education (years) ABC score ApoE genotype Sex
None 79 18 A0B1C0 23 Male
None 101 12 A1B1C0 23 Female
None 72 16 A1B0C0 33 Female
None 80 14 A1B1C0 33 Female
None 72 15 A2B2C0 34 Male
None 77 20 A3B3C3 34 Male
None 84 10 A3B3C3 33 Female
None 102 14 A3B2C1 33 Male
None 89 12 A3B2C1 34 Female
None 86 14 A3B2C2 33 Female
Mild 84 8 A3B3C3 34 Female
Mild 94 13 A3B3C1 33 Female
Mild 94 16 A3B3C3 33 Male
Mild 87 13 A3B3C2 34 Male
Mild 73 16 A3B3C3 N/A Female
Moderate 71 16 A3B3C3 44 Female
Moderate 83 18 A3B2C1 44 Female
Moderate 86 16 A3B3C2 34 Male
Moderate 77 15 A3B3C3 44 Female
Moderate 85 9 A3B3C3 34 Male
Severe 93 20 A3B3C3 34 Male
Severe 93 12 A3B3C2 33 Female
Severe 93 14 A3B3C3 33 Female
Severe 78 20 A3B3C3 24 Male
Severe 88 14 A3B3C3 44 Female
Open in a new tab

Note: ABC Score—A‐score: Aβ/amyloid plaque score (Thal phases) (Thal et al. 2002); B‐score: NFT score (Braak stage) (Braak and Braak 1991), (Braak et al. 2006); C‐score: Neuritic plaque score (CERAD) (Mirra et al. 1991).

2.2. Neuropathological Assessment

Formalin‐fixed autopsy tissue from the left hemisphere of the brain was processed for histochemical and immunohistochemical evaluation. For evaluation of AD pathology, tissue was stained for Aβ with 10D5 (Eli Lilly) and for tau with PHF‐1 (courtesy of Peter Davies), and assigned an ABC score by an expert neuropathologist (Thal et al. 2002; Braak and Braak 1991; Mirra et al. 1991). For evaluation of CAA pathology, tissue was stained with 10D5 (Eli Lilly), and sections of occipital cortex were semiquantitatively scored by an expert neuropathologist as none, mild, moderate, and severe CAA for both leptomeningeal and parenchymal vessels, considering multiple factors including density, circumferential accumulation, and dyshoric features. Sections from the right frozen hemibrain were stained with X34 dye (Millipore Sigma Cat. No. SML1954) immunofluorescence to evaluate side‐by‐side concordance. Representative images of each CAA category are provided in Figure 1.

FIGURE 1.

FIGURE 1

Open in a new tab

(A) Schematic representation of the methodology for the extraction of amyloid‐β (Aβ) from frozen occipital cortex tissue prior to immunoprecipitation‐mass spectrometry (IP/MS). (B) Schematic diagram of Aβ peptide sequence showing various N‐ and C‐terminal isoforms detected in Alzheimer's disease (AD) brains using mass spectrometry. (C) Nomenclature for the stereoisomers of Aβ1‐x with either or both Asp‐1 and Asp‐7 isomerized (D/L‐iso‐Asp) along with the native Aβ1‐x where both Asp‐1 and Asp‐7 are in L form. (D) Schematic diagram of the Aβ peptide sequence showing sites of common PTMs investigated in this report (top). Schematic representation of the mechanism of dehydration of L‐Asp (native form) and deamidation of L‐Asp resulting in a succinimide intermediate that subsequently leads to the isomerization/racemization after ring opening to D/L‐iso‐Asp and D‐Asp. Schematic diagram of the Aβ peptide sequence showing sites of common PTMs investigated in this report (bottom). (E) Representative immunohistochemistry of Aβ antibody (10D5) stained brain sections from neuropathologically assessed cases annotated as (i) CAA‐none, (ii) CAA‐mild, (iii) CAA‐moderate, and (iv) CAA‐severe, with increased magnification in (v‐viii). Insert in (viii) represents the double‐barreled morphology in severe CAA. (scale bar: 100 μm).

2.3. Isolation of Cerebral Blood Vessels and Soluble and Insoluble Fractionation

Cerebral vessels from the postmortem brain samples of AD patients and controls were isolated as described in previous studies (Tontsch and Bauer 1989). Briefly, 2 g of frozen occipital cortex tissue was homogenized in ice‐cold buffer solution (130 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl2, 1.2 mM KH2PO4, 1.2 mM MgSO4, 15 mM Hepes, 25 mM NaHCO3, 10 mM glucose, and 1 mM sodium pyruvate) at pH 7.4 (Mehla et al. 2021). Next, we added an equal volume of 26% dextran (Sigma‐Aldrich Cat. No. 31390‐25G) to the homogenate and centrifuged at 5800 × g for 15 min at 4°C, which separated the samples into three layers: vessel‐free parenchyma (upper), myelin/debris (middle), and pellet containing vessels (bottom). The upper and middle layers were discarded. Cerebral vessels from the bottom layer were washed three times with ice‐cold buffered solution and further purified by passing over a 100 μm nylon mesh strainer (Fisher Scientific Cat. No. 22‐363‐549). The filtrate was then centrifuged at 5000 × g for 10 min at 4°C to collect cerebrovessels in pellet form. Isolated cerebral vessels were then suspended in ice‐cold 1× RIPA lysis buffer (20 mM Tris‐HCl at pH 7.5, 150 mM NaCl, 1 mM Na2EDTA, 1 mM EGTA, 1% NP‐40, 1% sodium deoxycholate, 2.5 mM sodium pyrophosphate, 1 mM beta‐glycerophosphate, 1 mM Na3VO4, 1 μg/mL leupeptin and protease inhibitor cocktail) (Mehla et al. 2021), were sonicated, and centrifuged at 20 000 × g for 20 min at 4°C. Supernatant was collected as the soluble fraction and stored at −80°C for further experiments. The insoluble amyloid‐β fraction was prepared using formic acid (Thermo Fisher Cat. No. 243954) as described previously (Michno et al. 2019; Tomidokoro et al. 2010). Briefly, pellets obtained after extraction with RIPA buffer were washed once with PBS (Sigma‐Aldrich Cat. No. P3813‐10PAK), then resuspended in ice‐cold 70% formic acid for extraction of insoluble Aβ. The formic acid fraction was neutralized in 20 volumes of 1M Tris solution (Thermo Fisher Cat. No. 15567‐027). The protein concentration in the supernatant from the insoluble fraction was determined using the DC protein assay kit from Bio‐Rad (Cat. No. 5000001).

2.4. Parenchymal Plaque Extraction and Fractionation

Parenchymal plaque analysis was performed from two separate cohorts: n = 11 ad brains were analyzed for N‐terminal isomer distribution as described in (Mukherjee, Perez, Lago, et al. 2021) and n = 2 ad brains were analyzed for N‐ and C‐terminal isoform distribution according to the same protocol. Briefly, frozen temporal cortex tissue (~0.25 g) was homogenized in Tris‐buffered saline (TBS) at pH 8.5 containing 1:4 w/v protease inhibitors (Roche Cat. No. 11836145001) and ultracentrifuged at 100 000 × g for 30 min at 4°C to yield the soluble TBS fraction. The pellet was sequentially extracted with 100 mM Na2CO3 at pH 11 for peripheral membrane proteins (Na2CO3 fraction), urea–detergent buffer (7 M urea, 2 M thiourea, 4% CHAPS) for membrane‐associated proteins (urea–detergent fraction), and finally 70% formic acid for insoluble aggregates (FA fraction). Each supernatant was collected after ultracentrifugation, snap‐frozen in liquid N2, and stored at −80°C.

2.5. Immunoprecipitation/Mass Spectrometry

For the soluble and insoluble Aβ peptide analysis of CAA and insoluble parenchymal plaques from the n = 2 ad brain cohort, we used the HJ5.1 antibody (courtesy of Hong Jiang and David Holtzman) for immunoprecipitation (IP) of Aβ. Each brain sample was normalized to a total tissue weight of 10 μg and diluted to 500 μL using 0.01% human serum albumin (Sigma‐Aldrich Cat. No. D5322) in milli‐Q water before adding antibody coupled beads. 20 μL of recombinant 15 N Aβ internal standard (0.1 ng/μL Aβ1–42, 0.01 ng/μL Aβ1–40, 0.01 ng/μL Aβ1–43, and 0.01 ng/μL Aβ1–38 prepared in 0.1% NH4OH/20% ACN) was spiked into the samples. 50 μL of HJ5.1‐coupled DynaBeads (25 μg antibody/mg beads and ~18 μg antibody conjugated beads per IP, Thermo Fisher Cat. No. 14311D) and 25 μL of 1% IGEPAL (Sigma‐Aldrich Cat. No. I8896‐100ML), 1× protease inhibitor, 5 mM guanidine in PBS pH 7.4 were added to the samples. The IP and bead washing steps were performed in a KingFisher automated format; Aβ peptides were eluted off beads using neat formic acid (99% FA, Fisher Scientific Cat. No. A117‐50) and lyophilized to dryness. The samples were resuspended in 80 μL of 100 mM TEABC (Sigma‐Aldrich Cat. No. 17408‐500ML), and 20 μL of 2.5 ng/μL Lys‐N enzyme (ImmunoPrecise Cat. No. L101‐0.5 mg) was added for overnight digestion at 37°C. The following day the peptides were desalted using an Oasis HLB μElution plate (Waters Cat. No. 186001828BA) according to the manufacturer's protocol and lyophilized to dryness. During the desalting process, 10 fmol each of AQUA internal‐standard (Thermo Fisher) peptide for Aβ1–15, Aβ2–15, Aβ3pGlu‐15, and Aβ4–15 were spiked for validation based on retention time and matching MS2 transitions. The peptides were incubated in 3% H2O2 (Fisher Scientific Cat. No. H325‐500) and 3% FA for 18 h at 4°C to doubly oxidize the methionine within Aβ peptides. The following day, samples were desalted using an Oasis μElution HLB plate (Waters) using the manufacturer's protocol and lyophilized. The Aβ peptides were reconstituted in a 2% ACN/2% FA solution containing 20 fmol/μL BSA tryptic peptides for MS analysis on a Vanquish Neo UHPLC (Thermo Fisher Scientific) coupled to Orbitrap Exploris 480 (Thermo Fisher Scientific).

The peptides were directly loaded onto HSS T3 75 μm × 100 μm, 1.8 μm C18 column (Waters, heated to 65°C) using a Vanquish Neo UHPLC (Thermo Fisher Scientific) with a flow rate of 0.7 μL/min using buffer A (0.1% FA in water) and buffer B (0.1% FA in ACN, Fisher Scientific Cat. No. A955‐1). The Aβ peptides were eluted off the column using a flow rate of 0.4 μL/min with a gradient of buffer B of 0.5%–4% for 2 min, 4%–8% for 5 min, 8%–20% for 10 min, and 20%–35% for 3 min before ramping to 90% buffer B and cleaning the column for the next 3 min. The column was equilibrated back to 0.5% buffer B for the next 4 min before starting another run. The Orbitrap Exploris 480 (Thermo Fisher Scientific) was equipped with a Nanospray Flex electrospray ion source (Thermo Fisher Scientific) and operated in positive ion mode. Peptide ions were sprayed from a 10 μm SilicaTip emitter (New Objective) into the ion source (spray voltage = 2200 V, ion source temperature 275°C). Peptide precursor ions were targeted and isolated in the quadrupole. Isolated ions were fragmented by HCD, and fragment ions were detected in the Orbitrap (resolution of 30 000–60 000 for Aβ peptides, mass range 150–1500 m/z). More details regarding these transitions are provided in the Appendix S1. The raw MS data were imported into Skyline (University of Washington, v4.2) with formula annotations of the targeted peptides. Aβ peptides (Aβ1–15, Aβ2–15, Aβ4–15, Aβ3pGlu‐15, Aβ16–27, Aβ28–37, Aβ28–38, Aβ28–39, Aβ28–40, Aβ28–41, Aβ28–42, and Aβ28–43) were quantified by comparison with the corresponding isotopomer signals from their respective 15 N internal standard (15 N Aβ1–15 peak area was used to quantitate Aβ2–15, Aβ4–15, and Aβ3pGlu‐15, while 15 N Aβ28–38, 15 N Aβ28–40, and 15 N Aβ28–42 and were used to quantitate Aβ28–37, Aβ28–39, and Aβ28–41, respectively).

2.6. Statistical Analysis

Data analyses were performed in Tableau, GraphPad Prism 10, and R. Statistical analysis of the Aβ isomers, isoforms, and their fractional values of each group were calculated using the nonparametric Mann–Whitney test and Kruskal–Wallis test followed by post hoc pairwise multiple comparisons using the two‐stage method of Benjamini, Kreiger, and Yekutieli. p values were adjusted for multiple comparison false discovery rate (p < 0.05). For correlation heatmaps of Aβ species, nonparametric Spearman correlations were performed (confidence interval of 95%). The data in this study are not assessed for normality, and no outliers were removed.

2.7. Ethics and Consent to Participate Declarations

Autopsy samples were obtained from the Charles F. and Joanne Knight Alzheimer's Disease Research Center at Washington University in St. Louis. All participants gave prospective premortem written consent for their brains to be banked and used for research.

3. Results

To investigate biochemical differences in Aβ composition across CAA severity and parenchymal plaques, we quantified both the absolute concentrations and relative abundances of various Aβ species. Targeted mass spectrometry (MS/MS) analysis was performed on insoluble and soluble fractions of purified vascular extracts from mild, moderate, and severe CAA and controls without CAA. Proteoform distributions were then compared across CAA severities, to controls without CAA, and to parenchymal Aβ plaques. In most cases, the insoluble and soluble vascular fractions (Figure S1) showed similar proteoform distributions.

3.1. Distinct N‐Terminal and C‐Terminal Isoforms in CAA and Parenchymal Plaques

To assess differences in Aβ N‐terminal truncation, we first analyzed the insoluble fraction from CAA‐enriched vessels. Aβ1–15 was the primary N‐terminal isoform detected (Figure 2Ai), with significantly higher concentrations in moderate and severe CAA compared to CAA‐none controls. In contrast, the mild CAA group was not significantly different from controls. Quantitation of additional N‐terminal isoforms Aβ2–15, Aβ4–15, Aβ3pGlu‐15, cit‐Aβ3pGlu‐15 showed concentrations over 100 times lower than Aβ1–15 in CAA‐positive vessels. These isoforms followed a similar pattern wherein only moderate and severe CAA showed significantly higher levels than the control CAA‐none group, with the exception of cit‐Aβ3pGlu‐15, which did not show any significant differences across mild, moderate, and severe CAA (Figure 2Aii–v).

FIGURE 2.

FIGURE 2

Open in a new tab

Distinct N‐terminal and C‐terminal Aβ proteoforms distributions in CAA and parenchymal plaques. Box plots representing the vascular insoluble Aβ isoform distribution of (A) N‐terminal Aβ proteoforms (i) Aβ1–15, (ii) Aβ2–15, (iii) Aβ3pGlu‐15, (iv) cit‐Aβ3pGlu‐15, and (v) Aβ4–15, (B) C‐terminal Aβ proteoforms (i) Aβ37, (ii) Aβ38, (iii) Aβ39, (iv) Aβ40, (v) Aβ41, (vi) Aβ42, and (vii) Aβ43, and (C) Aβ1–15 isomers (Asp‐1, Asp‐7) in parenchymal plaques and cerebral amyloid angiopathy (CAA). The p values were FDR adjusted (p < 0.05); statistical significance was determined by using a two‐tailed nonparametric Kruskal–Wallis test (p < 0.05), (n = 25 participants in A‐B and n = 36 participants in C).

We next investigated C‐terminal truncations of Aβ in the vascular insoluble fraction. The concentrations of Aβ37, Aβ38, Aβ39, and Aβ40 were markedly elevated in moderate and severe CAA (Figure 2Bi–iv). These effects were absent in the mild CAA subgroup, which was not different from controls. Among C‐terminal peptides, Aβ40 was the dominant species whose concentrations in moderate and severe CAA were 2–10 times higher compared with other C‐terminal forms. Aβ41 was detected at low levels (less than 0.1%) and showed similar trends to the aforementioned C‐terminal peptides. In contrast, concentrations of Aβ42 and Aβ43 do not show the same large increase with moderate or severe CAA (Figure 2Bvi,vii).

3.2. Contrasting Aβ1–15 Isomerization (Asp‐1, Asp‐7) in Parenchymal Plaques and CAA

We examined the distribution of Aβ1‐15 isomers in parenchymal plaques and CAA. Parenchymal Aβ plaques, vessels without CAA, and vessels with mild CAA showed low levels (~15%) of native Aβ1–15 (Asp1,7L). In contrast, moderate and severe CAA showed a marked increase in native Aβ1–15 (Figure 2Ci), which reached statistical significance for the moderate group only. Quantification of the additional isomers revealed unique distributions in Aβ plaques, vessels with none or mild CAA, and vessels with moderate or severe CAA. The mono‐iso‐L‐Asp form was relatively stable at ~25% across all conditions (Figure 2Cii). The di‐iso‐L‐Asp form was the most abundant isomer found in Aβ parenchymal plaques and vessels with none or mild CAA, where it accounts for ~45% of the Aβ1‐15 (Figure 2Ciii). However, moderate and severe CAA showed lower representation of this isoform (Figure 2Civ). The percentage of mono‐iso‐D‐di‐iso‐Asp found in CAA‐none and CAA‐mild was ~25%. In contrast, parenchymal Aβ plaques and moderate CAA showed lower representation (15%).

3.3. Distribution of Aβ N‐Terminus Isoforms in Insoluble Fractions of (a) CAA and (b) Parenchymal Aβ Plaques

To ensure all major N‐terminal isoforms were accounted for and to investigate the contribution of each isoform across severities of CAA pathology and in parenchymal plaques, we calculated the fractional abundance of each isoform as a percentage of all isoforms measured. In concordance with the absolute quantitation results above, the fraction of N‐terminal Aβ species in CAA was dominated by Aβ1–15 representing 94.6% in mild, 98.2% in moderate, and 97.3% in severe conditions (Figure 3Ai). In mild CAA, the rank order of fractional abundance of N‐terminal species was Aβ1–15 >> Aβ4–15 > Aβ3pGlu > Aβcit‐3pGlu > Aβ2–15. In moderate CAA, the order shifted slightly to: Aβ1–15 >> Aβ3pGlu‐15 > Aβ4–15 > Aβ2–15 > Aβcit‐3pGlu‐15. In severe CAA, the distribution was Aβ1–15 >> Aβ4–15 > Aβ3pGlu‐15 > Aβ2–15 > Aβcit‐3pGlu‐15 (see Table S1 for percentages fractional abundance of Aβ isoforms). The control CAA‐none group, which had overall very low absolute Aβ levels, showed a distribution most similar to mild CAA. In contrast, parenchymal plaques displayed greater heterogeneity in N‐terminal isoform composition (Figure 3Aii). Aβ1–15 comprised around 48% of the N‐terminal species, followed by Aβ3pGlu‐15 at ~24%, Aβ2–15 at ~14%, and Aβ4–15 at ~13%. The relatively high abundance of Aβ3pGlu‐15 in our results aligns with previous findings in AD plaques (Roher, Lowenson, Clarke, Woods, et al. 1993), where this modification was among the predominant forms. Overall, the fractional abundance of N‐terminal species in parenchymal plaques exhibits far greater heterogeneity compared to CAA, with the rank order: Aβ1–15 > Aβ3pGlu > Aβ2–15 > Aβ4–15.

3.4. Distribution of Aβ C‐Terminus Isoforms in Insoluble Fractions of (a) CAA and (b) Parenchymal Aβ Plaques

For the fractional C‐terminal Aβ species, we detected a larger proportion of Aβ42, approximately 78% in mild CAA compared to about 25% in moderate and 25% in severe CAA (Figure S2). In contrast, the proportion of Aβ40 in mild CAA was ~17%, while that in moderate and severe CAA was ~53% and ~58%, respectively (Figure 3Bi). Again, the CAA‐none group had low absolute levels and a distribution most closely aligned with mild CAA. The fractional C‐terminal Aβ distribution species from the insoluble fractions is presented in Figure S3A. With increasing severity of CAA, the dominant C‐terminal species shifted from Aβ42 to Aβ40. The rank order of C‐terminal isoforms in mild CAA was: Aβ42 > Aβ40 > Aβ43 > Aβ38 > Aβ39 > Aβ37 > Aβ41; in moderate CAA: Aβ40 > Aβ42 > Aβ38 > Aβ39 > Aβ37 > Aβ43 > Aβ41; and in severe CAA: Aβ40 > Aβ42 > Aβ38 > Aβ43 > Aβ39 > Aβ37 > Aβ41. In parenchymal plaques, Aβ42 overwhelmingly dominated the C‐terminal profile, accounting for 96% of total signal. The remaining ~4% consisted of other C‐terminal species in the following decreasing order: Aβ43 > Aβ40 > Aβ38 > Aβ41 > Aβ37 > Aβ39 (Figure 3Bii). These findings highlight the greater C‐terminal heterogeneity in moderate and severe CAA, characterized by an increased presence of truncated, less hydrophobic Aβ species such as Aβ37, Aβ38, Aβ39, and Aβ40, whereas Aβ42 remains the predominant C‐terminus in parenchymal plaques.

FIGURE 3.

FIGURE 3

Open in a new tab

Heterogeneity of insoluble Aβ isoforms in CAA and parenchymal Aβ plaques. Box plots illustrating the fractional abundance of (A) N‐terminal Aβ isoforms, (B) C‐terminal Aβ isoforms and (C) correlations between N‐ and C‐terminal Aβ species for CAA Aβ positive samples, where (Ci) shorter C‐terminal species Aβ37, Aβ38, and Aβ39 strongly correlated with each other and with native Aβ1–15 but showed negative correlations with Aβ42 and Aβ43. Aβ42 and Aβ43 were highly correlated with each other. In parenchymal plaques (Cii) Aβ40 and Aβ42 were inversely correlated. Together these results reveal distinct isoform distribution and opposing N‐ and C‐terminal relationships in vascular and parenchymal Aβ deposits. Spearman's rank correlation was used for all correlation analyses. (n = 25 participants in Ai, Bi; n = 2 participants in Aii and Bii, n = 15 participants in Ci and n = 11 participants in Cii).

3.5. Correlation of Aβ C‐Terminus and N‐Terminus in CAA and Parenchymal Aβ Plaques

In mild, moderate, and severe CAA, the relative proportions of shorter C‐terminal Aβ species Aβ37, Aβ38, Aβ39, and Aβ40 exhibited strong positive correlations with each other (r = 0.58–1.00, p = 0.021 to 2.29 × 10−11) (Figure 3Ci and Figure S3). Additionally, the longer C‐terminal Aβ species Aβ42 and Aβ43 were highly correlated (r = 0.99, p < 0.0001). In contrast, the shorter C‐terminal species show a significant negative correlation (r = −0.80 to 0.85, p = 0.0002 to 0.0006) to Aβ42 and Aβ43 peptides. Additionally, our correlational analysis demonstrated that the shorter C‐terminal peptides Aβ37, Aβ38, and Aβ39 exhibited strong positive correlation with native Aβ1–15 (r = 0.96, 0.95, and 0.96 respectively, p = 1.24–2.8 × 10−7). Aβ40 showed a weaker but still significant correlation with native Aβ1–15 (r = 0.54, p = 0.04). Furthermore, we also observed a significant positive correlation between di‐iso‐Asp‐D‐Aβ1–15 and Aβ42 (r = 0.67, p value = 0.008), as well as Aβ43 (r = 0.62, p value = 0.01). Although similar trends were observed between mono‐iso‐Asp‐Aβ1–15 and mono‐D‐di‐iso‐Asp‐Aβ1–15, these did not reach statistical significance. Conversely, isomerized N‐terminal isoforms (di‐iso‐Asp‐L‐Aβ1–15 and mono‐iso‐D‐di‐iso‐Asp‐Aβ1‐15) showed significant negative correlations with the shorter C‐terminal isoforms Aβ37, Aβ38, and Aβ39 (r = −0.71 to −0.86, p = 0.004–0.0001). In parenchymal Aβ plaques, we observed a strong inverse correlation between Aβ40 and Aβ42 (r = −1.00, p = 6 × 10−7) (Figure 3Cii). We also observed a significant negative correlation (r = −0.77 to −0.90, p = 0.001 to 1 × 10−5) between the native N‐terminus Asp1,7L (native Aβ1–15) and the N‐terminal Asp isomers (di‐iso‐Asp‐Aβ1–15 and mono‐D‐di‐iso‐Asp‐Aβ1–15). The isomerized N‐terminal species mono‐D‐di‐iso‐Asp‐Aβ1–15 show positive correlation with di‐iso‐Asp‐Aβ1–15 (r = 0.77, p = 0.007), and di‐iso‐Asp‐D‐Aβ1–15 (r = 0.90, p = 0.0003).

4. Discussion

By employing a vessel purification strategy, coupled with Lys‐N digested Aβ‐targeted immunoprecipitation/mass spectrometry, we were able to characterize the distinct Aβ proteoform signature of CAA, how it changes with disease severity, and how it compares to Aβ profiles in parenchymal plaques of AD. Previous work using mass spectrometry has established that Aβ40 predominates in CAA, whereas Aβ42 is the dominant isoform in parenchymal amyloid plaques of AD (Michno et al. 2019, 2022; Mukherjee, Perez, Lago, et al. 2021; Rostagno et al. 2022; Koutarapu et al. 2025). Our findings are consistent with this established pattern and extend current understanding by providing the first quantitative analysis of N‐terminal and C‐terminal isoforms, along with key PTMs, across a graded spectrum of CAA severity.

One of the main findings of our study is that the vast majority of vascular insoluble Aβ exists without any N‐terminal truncation in moderate and severe CAA. This contrasts with parenchymal plaque Aβ, where only 50% of peptides retain the full N‐terminus, and truncated forms such as Aβ2‐x, Aβ4‐x, and Aβ3pGlu are more prevalent. We also quantified Aβ isomerization at Asp1 and Asp7, two spontaneous PTMs previously linked to Aβ aggregate aging in plaques. Compared to parenchymal plaques, CAA aggregates exhibited reduced spontaneous isomerization at both sites. Although the precise mechanism for this differential processing remains to be elucidated, the distinct structural constraints of vascular versus parenchymal Aβ fibrils may offer an explanation, as revealed by recent cryo‐electron microscopy (cryo‐EM) structures of ex vivo Aβ protofibrils extracted from AD patients.

The structural differences of the protofibril folds of Aβ in CAA versus parenchymal plaques, as revealed by cryo‐EM, suggest that modifications of the N‐terminus might be governed by accessibility. In parenchymal plaques, Aβ peptides adopt an S‐shaped fold such that the N‐terminus remains unstructured as part of the fuzzy coat; the disordered N‐terminus extends from Asp1‐Gly9 (Yang et al. 2022). In contrast, the vascular peptide fold is C‐shaped, with its N‐ and C‐termini forming arches and the peptide chain folding back onto the central peptide domain (Kollmer et al. 2019). Due to the involvement of the N‐terminus in the peptide fold, it follows that solvent accessibility to the N‐terminus would be limited in the vascular compartment resulting in less PTMs as compared to parenchymal plaques.

While our data recapitulate prior work showing that Aβ40 is the dominant C‐terminal isoform in CAA (Kasri et al. 2024; Koutarapu et al. 2025; Michno et al. 2022), we have also measured shorter, less hydrophobic Aβ37, Aβ38, Aβ39, along with rarer C‐terminal fragments like Aβ41 in trace quantities, the pathological significance of which remains to be elucidated. In characterizing the distribution of C‐terminal fragments, we observed a progressive decrease in the fractional abundance of longer, more hydrophobic C‐terminal species such as Aβ42 and Aβ43, and a corresponding increase in shorter, less hydrophobic species such as Aβ37, Aβ38, Aβ39, and Aβ40 with increasing CAA severity. This pattern suggests that hydrophobic C‐terminal isoforms, particularly Aβ42, might be more functionally relevant to the aggregation process early in the disease course, with Aβ40 and other shorter C‐terminal fragments becoming dominant later with disease progression in the vasculature. This shift could be compatible with the hypothesis that Aβ42 serves as a nidus for the aggregation of Aβ40 in the vessels (Sowade and Jahn 2017). These results highlight that C‐terminal heterogeneity is another key distinguishing feature for CAA compared to parenchymal plaques (Fulcher et al. 2025). However, how this proteoform diversity in the vasculature and parenchyma impacts anti‐amyloid therapies remains largely unexplored. Critically, it is still unclear which vascular Aβ proteoforms are recognized and targeted by the currently approved anti‐Aβ monoclonal antibodies that may potentially lead to treatment‐related complications. Adverse side effects such as ARIA continue to pose major challenges to safe and effective treatment for AD. Our analytical platform offers a valuable tool for elucidating the molecular mechanisms underlying ARIA, including the potential involvement of specific vascular Aβ species, and may enable the development of less‐invasive biomarkers for stratifying patients at higher risk.

Our combined IP/MS technique expands and complements information provided by structural methods (cryo‐EM) and traditional immunohistochemical studies to further delineate the mechanism by which the same peptide can form distinct fibrils in these two brain compartments. By using mass spectrometry to characterize the proteoform distributions, we have circumvented many of the limitations of immunohistochemistry alone, such as the need for antibodies specific to each truncation or modification. While proteomic investigations generally lose structural information regarding the large, folded proteins following enzymatic digestion, our strategy of probing the structural PTMs (Asp residue isomerization) provides new insights into Aβ protofibril folds in distinct brain compartments. Disordered regions like the “fuzzy” coat of the aggregates are often unresolved by cryo‐EM, particularly for many disease‐associated proteins (Yang et al. 2023; Scheres et al. 2023; Lövestam et al. 2024), underscoring the value of biochemical approaches that probe proteoform architecture at the residue level.

We observed a largely binary response with none and mild‐CAA showing similar distributions of Aβ proteoforms, while moderate and severe‐CAA showed similar distributions. Interestingly, most of the statistically significant differences were present in only the moderate CAA cohort and did not reach significance for the severe cohort. This may be secondary to the regional variability in CAA pathology within brain regions such that tissue submitted for IP/MS analysis may have incompletely represented the pathology present in adjacent regions evaluated by routine IHC for neuropathologic grading (Attems 2005). Another possible explanation is the enrichment of APOE4 carriers within the moderate group (Table 1). There is a well‐known relationship between APOE4 and Aβ levels (Fleisher et al. 2013; Vemuri et al. 2010), which we recapitulate in our data, wherein APOE4 carriers show significantly higher vascular levels of all isoforms than noncarriers (Figure S5). APOE4 carriers show lower fractional abundance of hydrophobic isoforms Aβ42 and Aβ43 and higher fractional abundance of the shorter Aβ isoforms, reflecting their enrichment in the moderate and severe CAA groups (Figure S6). Given the known status of APOE4 as a risk factor CAA (Greenberg et al. 1995; Rannikmäe et al. 2014; Shinohara et al. 2016) and recent work demonstrating effects of APOE on Aβ clearance (Xia et al. 2024), future investigation on if and how APOE4 is driving the isoform shifts measured in this work may provide mechanistic insight.

While this study provides novel insights into Aβ proteoform distributions in vascular and parenchymal aggregates, we acknowledge several important limitations. Our vessel purification protocol excludes larger vessels such as leptomeningeal arteries, limiting the scope of vascular compartments analyzed. Additionally, all tissues were derived from the occipital cortex, which may not fully reflect regional heterogeneity across the brain. Future work using selective capture of vessels stratified by location (leptomeningeal vs. parenchymal), size (arterioles vs. capillaries), or morphology (non‐circumferential vs. circumferential vs. dyshoric) will better characterize vessel type‐specific proteoform distributions across multiple brain regions. Mass spectrometry imaging approaches such as MALDI‐MS imaging may also allow spatially resolved comparisons of N‐ and C‐terminal isoforms across vessel types and regions (Michno et al. 2022). Finally, broadening the current targeted Aβ proteomics to an untargeted proteomic approach could uncover additional interacting proteins and biological pathways, offering a more comprehensive view of disease mechanisms.

5. Conclusions

In summary, we provide the first systematic characterization of Aβ proteoform distributions across graded severities of CAA, comparing them directly to those found in parenchymal plaques. We show that vascular Aβ in CAA exhibits N‐terminal uniformity with minimal Asp isomerization, in contrast to the extensive N‐terminal modifications observed in parenchymal plaques, including Asp isomerization, pyroglutamate formation, and sequential truncations. Additionally, we identify C‐terminal heterogeneity as a defining feature of CAA, marked by enrichment in shorter, less hydrophobic Aβ isoforms (Aβ37–Aβ40), whereas Aβ42 predominates in parenchymal aggregates. We propose that these differences arise from the structural variations in Aβ protofibril folds, which influence accessibility and susceptibility of specific regions of peptide to PTMs (Figure 4). Our findings offer insight into compartment‐specific Aβ aggregation pathways that could be useful in designing targeted therapeutics against distinct Aβ species. Together, our results advance the molecular understanding of Aβ aggregation in CAA and AD and highlight new avenues for both diagnostic and therapeutic development.

FIGURE 4.

FIGURE 4

Open in a new tab

Differences of the protofibril folds of amyloid‐β (Aβ) in cerebral amyloid angiopathy (CAA) versus parenchymal plaques. Dense cored plaque, diffuse plaques, and neuritic plaques in the parenchyma primarily contain protofilaments (red) consisting of G9‐A42 or V12‐A42 folds (gray ribbon) with the N‐terminus being highly disordered (pink dashed region) and prone to post‐translational modifications (PTMs) such as Asp isomerization of Asp1/Asp7, while vascular protofibrils (red) contain D1‐G38 folds with some disorder in the C‐terminus and an ordered N‐terminus. Unmodified (native) N‐terminus is a key feature of the vascular amyloid in CAA. We hypothesize that Aβ isoform and PTM heterogeneity are thus a consequence of the structure of the protofilaments that are formed in these two distinct aggregates.

Author Contributions

Srinivas Koutarapu: formal analysis, writing – original draft, visualization. Kaleigh F. Roberts: formal analysis, writing – original draft. Reid A. Coyle: investigation. Jogender Mehla: investigation, writing – review and editing, supervision. Chihiro Sato: writing – review and editing, resources. Gregory J. Zipfel: writing – review and editing, supervision. Randall J. Bateman: writing – review and editing, resources, supervision. Katherine E. Schwetye: conceptualization, writing – review and editing, funding acquisition, supervision. Soumya Mukherjee: conceptualization, investigation, formal analysis, writing – review and editing, funding acquisition, resources, supervision, data curation.

Funding

This study was supported by resources and effort provided by the Tracy Family Stable Isotope Labeling Quantitation Center (Principal Investigator (PI) R.J.B.) established by the Tracy Family, Richard Frimel & Gary Werths, GHR Foundation, Pat and Jane Tracy, Anonymous, Anne & Ray Capestrain, Community Foundation Serving West Central Illinois and Northeast Missouri, JTL Family Fund, Payne Family, Mary & Jay Sullivan, Tracy Family Foundation, Catherine & Tom Tracy, Community Foundation for the Land of Lincoln, Jim & Jil Tracy, Joe & Jill Tracy, Sonja & Robert M. Willman, Boniface Foundation, Jean & Michael Buckley, Ann Liberman, Clemence S. Lieber Foundation, Mary Schoolman & Dr. James Hinrichs, and Susan & Scott Stamerjohn brought together by The Foundation for Barnes‐Jewish Hospital. Funding for this work includes pilot grant from the Tracy Family SILQ center (K.E.S. and S.M.), with support from the National Institute on Aging (Healthy Aging and Senile Dementia) [P01 AG003991], National Institute of Aging (Knight Alzheimer's Disease Research Center (ADRC)) [P30 AG066444], National Institute of Aging (Adult Children Study) [P01 AG026276], NIH [NS103276 (G.J.Z), NS071011 (G.J.Z)], Coins for Alzheimer's Research Trust (CART) grant (C.S.) (Horie et al. 2020). This work was also supported by cores, resources and access to equipment made possible by the Hope Center for Neurological Disorders and the Department of Neurology at Washington University School of Medicine.

Conflicts of Interest

Washington University and R.J.B. have an equity ownership interest in C2N Diagnostics and receive income based on technology licensed by Washington University to C2N Diagnostics. R.J.B. receives income from C2N Diagnostics for serving on the scientific advisory board. R.J.B. has received research funding from Avid Radiopharmaceuticals, Janssen, Roche/Genentech, Eli Lilly, Eisai, Biogen, AbbVie, Bristol Myers Squibb, and Novartis. R.J.B. serves as an unpaid member on scientific advisory boards for Roche and Biogen. The remaining authors declare no competing interests.

Supporting information

Appendix S1: jnc70335‐sup‐0001‐AppendixS1.pdf.

JNC-169-0-s001.pdf (1.5MB, pdf)

Acknowledgments

We thank the participants and personnel of the Knight Alzheimer Disease Research Center, as well as the staff of Washington University's Translational Human Neurodegenerative Disease Research (THuNDR) Laboratory for providing postmortem human brain tissue for this study. We thank Vitaliy Ovod for helping in data analysis on Tableau software.

Koutarapu, S. , Roberts K. F., Coyle R. A., et al. 2025. “Post‐Translational Modifications Distinguish Amyloid‐β Isoforms in Cerebral Amyloid Angiopathy and Alzheimer's Disease.” Journal of Neurochemistry 169, no. 12: e70335. 10.1111/jnc.70335.

Srinivas Koutarapu and Kaleigh F. Roberts shared equal contribution.

Jogender Mehla, Katherine E. Schwetye, and Soumya Mukherjee are senior authors.

Contributor Information

Katherine E. Schwetye, Email: schwetyk@wustl.edu.

Soumya Mukherjee, Email: msoumya@wustl.edu.

Data Availability Statement

The quantitative data is available as Appendix S1 accompanying this article. Additional data that support the findings of this study are available from the corresponding author upon reasonable request. A preprint of this article was posted on BioRxiv on June 15, 2025 (doi: https://doi.org/10.1101/2025.06.11.658821).

References

  1. Attems, J. 2005. “Sporadic Cerebral Amyloid Angiopathy: Pathology, Clinical Implications, and Possible Pathomechanisms.” Acta Neuropathologica 110: 345–359. [DOI] [PubMed] [Google Scholar]
  2. Attems, J. , Jellinger K., Thal D. R., and Van Nostrand W.. 2011. “Review: Sporadic Cerebral Amyloid Angiopathy.” Neuropathology and Applied Neurobiology 37: 75–93. [DOI] [PubMed] [Google Scholar]
  3. Braak, H. , Alafuzoff I., Arzberger T., Kretzschmar H., and Del Tredici K.. 2006. “Staging of Alzheimer Disease‐Associated Neurofibrillary Pathology Using Paraffin Sections and Immunocytochemistry.” Acta Neuropathologica 112: 389–404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Braak, H. , and Braak E.. 1991. “Neuropathological Stageing of Alzheimer‐Related Changes.” Acta Neuropathologica 82: 239–259. [DOI] [PubMed] [Google Scholar]
  5. Cabrera, E. , Mathews P., Mezhericher E., et al. 2018. “Aβ Truncated Species: Implications for Brain Clearance Mechanisms and Amyloid Plaque Deposition.” Biochimica et Biophysica Acta (BBA) ‐ Molecular Basis of Disease 1864: 208–225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Crehan, H. , Liu B., Kleinschmidt M., et al. 2020. “Effector Function of Anti‐Pyroglutamate‐3 Aβ Antibodies Affects Cognitive Benefit, Glial Activation and Amyloid Clearance in Alzheimer's‐Like Mice.” Alzheimer's Research & Therapy 12: 12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Fleisher, A. S. , Chen K., Liu X., et al. 2013. “Apolipoprotein E ε4 and Age Effects on Florbetapir Positron Emission Tomography in Healthy Aging and Alzheimer Disease.” Neurobiology of Aging 34: 1–12. [DOI] [PubMed] [Google Scholar]
  8. Fossati, S. , Todd K., Sotolongo K., Ghiso J., and Rostagno A.. 2013. “Differential Contribution of Isoaspartate Post‐Translational Modifications to the Fibrillization and Toxic Properties of Amyloid β and the Asn23 Iowa Mutation.” Biochemical Journal 456: 347–360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Fulcher, J. M. , Ives A. N., Tasaki S., et al. 2025. “Discovery of Proteoforms Associated With Alzheimer's Disease Through Quantitative Top‐Down Proteomics.” Molecular & Cellular Proteomics 24: 100983. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Gkanatsiou, E. , Portelius E., Toomey C. E., et al. 2019. “A Distinct Brain Beta Amyloid Signature in Cerebral Amyloid Angiopathy Compared to Alzheimer's Disease.” Neuroscience Letters 701: 125–131. [DOI] [PubMed] [Google Scholar]
  11. Goedert, M. , Spillantini M. G., Jakes R., Rutherford D., and Crowther R. A.. 1989. “Multiple Isoforms of Human Microtubule‐Associated Protein Tau: Sequences and Localization in Neurofibrillary Tangles of Alzheimer's Disease.” Neuron 3: 519–526. [DOI] [PubMed] [Google Scholar]
  12. Greenberg, S. M. , Rebeck G. W., Vonsattel J. P., Gomez‐Isla T., and Hyman B. T.. 1995. “Apolipoprotein E Epsilon 4 and Cerebral Hemorrhage Associated With Amyloid Angiopathy.” Annals of Neurology 38: 254–259. [DOI] [PubMed] [Google Scholar]
  13. He, W. , and Barrow C. J.. 1999. “The Aβ 3‐Pyroglutamyl and 11‐Pyroglutamyl Peptides Found in Senile Plaque Have Greater β‐Sheet Forming and Aggregation Propensities In Vitro Than Full‐Length Aβ.” Biochemistry 38: 10871–10877. [DOI] [PubMed] [Google Scholar]
  14. Horie, K. , Barthélemy N. R., Mallipeddi N., et al. 2020. “Regional Correlation of Biochemical Measures of Amyloid and Tau Phosphorylation in the Brain.” Acta Neuropathologica Communications 8: 149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Kakuda, N. , Miyasaka T., Iwasaki N., et al. 2017. “Distinct Deposition of Amyloid‐β Species in Brains With Alzheimer's Disease Pathology Visualized With MALDI Imaging Mass Spectrometry.” Acta Neuropathologica Communications 5: 73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kasri, A. , Camporesi E., Gkanatsiou E., et al. 2024. “Amyloid‐β Peptide Signature Associated With Cerebral Amyloid Angiopathy in Familial Alzheimer's Disease With APPdup and Down Syndrome.” Acta Neuropathologica 148: 8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Keage, H. A. D. , Carare R. O., Friedland R. P., et al. 2009. “Population Studies of Sporadic Cerebral Amyloid Angiopathy and Dementia: A Systematic Review.” BMC Neurology 9: 3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kollmer, M. , Close W., Funk L., et al. 2019. “Cryo‐EM Structure and Polymorphism of Aβ Amyloid Fibrils Purified From Alzheimer's Brain Tissue.” Nature Communications 10: 4760. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Koutarapu, S. , Ge J., Dulewicz M., et al. 2025. “Chemical Imaging Delineates Aβ Plaque Polymorphism Across the Alzheimer's Disease Spectrum.” Nature Communications 16: 3889. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kuo, Y.‐M. , Emmerling M. R., Woods A. S., Cotter R. J., and Roher A. E.. 1997. “Isolation, Chemical Characterization, and Quantitation of Ab 3‐Pyroglutamyl Peptide From Neuritic Plaques and Vascular Amyloid Deposits.” Biochemical and Biophysical Research Communications 237: 188–191. [DOI] [PubMed] [Google Scholar]
  21. Lambeth, T. R. , Riggs D. L., Talbert L. E., et al. 2019. “Spontaneous Isomerization of Long‐Lived Proteins Provides a Molecular Mechanism for the Lysosomal Failure Observed in Alzheimer's Disease.” ACS Central Science 5: 1387–1395. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lövestam, S. , Li D., Wagstaff J. L., et al. 2024. “Disease‐Specific Tau Filaments Assemble via Polymorphic Intermediates.” Nature 625: 119–125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lyons, B. , Friedrich M., Raftery M., and Truscott R.. 2016. “Amyloid Plaque in the Human Brain Can Decompose From Aβ(1‐40/1‐42) by Spontaneous Nonenzymatic Processes.” Analytical Chemistry 88: 2675–2684. [DOI] [PubMed] [Google Scholar]
  24. Masters, C. L. , Simms G., Weinman N. A., Multhaup G., McDonald B. L., and Beyreuther K.. 1985. “Amyloid Plaque Core Protein in Alzheimer Disease and Down Syndrome.” Proceedings of the National Academy of Sciences of the United States of America 82: 4245–4249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Mehla, J. , Singh I., Diwan D., et al. 2021. “STAT3 Inhibitor Mitigates Cerebral Amyloid Angiopathy and Parenchymal Amyloid Plaques While Improving Cognitive Functions and Brain Networks.” Acta Neuropathologica Communications 9: 193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Michno, W. , Koutarapu S., Camacho R., et al. 2022. “Chemical Traits of Cerebral Amyloid Angiopathy in Familial British‐, Danish‐, and Non‐Alzheimer's Dementias.” Journal of Neurochemistry 163: 233–246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Michno, W. , Nyström S., Wehrli P., et al. 2019. “Pyroglutamation of Amyloid‐βx‐42 (Aβx‐42) Followed by Aβ1–40 Deposition Underlies Plaque Polymorphism in Progressing Alzheimer's Disease Pathology.” Journal of Biological Chemistry 294: 6719–6732. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Miller, D. L. , Papayannopoulos I. A., Styles J., et al. 1993. “Peptide Compositions of the Cerebrovascular and Senile Plaque Core Amyloid Deposits of Alzheimer's Disease.” Archives of Biochemistry and Biophysics 301: 41–52. [DOI] [PubMed] [Google Scholar]
  29. Mintun, M. A. , Lo A. C., Duggan Evans C., et al. 2021. “Donanemab in Early Alzheimer's Disease.” New England Journal of Medicine 384: 1691–1704. [DOI] [PubMed] [Google Scholar]
  30. Mirra, S. S. , Heyman A., McKeel D., et al. 1991. “The Consortium to Establish a Registry for Alzheimer's Disease (CERAD): Part II. Standardization of the Neuropathologic Assessment of Alzheimer's Disease.” Neurology 41: 479. [DOI] [PubMed] [Google Scholar]
  31. Mukherjee, S. , Coyle R., Dubois C., et al. 2025. “Isomerized Aβ in the Brain Can Distinguish the Status of Amyloidosis in the Alzheimer's Disease Spectrum.” Acta Neuropathologica 150: 7. [DOI] [PubMed] [Google Scholar]
  32. Mukherjee, S. , Perez K. A., Dubois C., et al. 2021. “Citrullination of Amyloid‐β Peptides in Alzheimer's Disease.” ACS Chemical Neuroscience 12: 3719–3732. [DOI] [PubMed] [Google Scholar]
  33. Mukherjee, S. , Perez K. A., Lago L. C., et al. 2021. “Quantification of N‐Terminal Amyloid‐β Isoforms Reveals Isomers Are the Most Abundant Form of the Amyloid‐β Peptide in Sporadic Alzheimer's Disease.” Brain Communications 3: fcab028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Portelius, E. , Bogdanovic N., Gustavsson M. K., et al. 2010. “Mass Spectrometric Characterization of Brain Amyloid Beta Isoform Signatures in Familial and Sporadic Alzheimer's Disease.” Acta Neuropathologica 120: 185–193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Rannikmäe, K. , Kalaria R. N., Greenberg S. M., et al. 2014. “APOE Associations With Severe CAA‐Associated Vasculopathic Changes – Collaborative Meta‐Analysis.” Journal of Neurology, Neurosurgery & Psychiatry 85: 300–305. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Reinert, J. , Richard B. C., Klafki H. W., et al. 2016. “Deposition of C‐Terminally Truncated Aβ Species Aβ37 and Aβ39 in Alzheimer's Disease and Transgenic Mouse Models.” Acta Neuropathologica Communications 4: 24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Roher, A. E. , Lowenson J. D., Clarke S., et al. 1993. “Structural Alterations in the Peptide Backbone of Beta‐Amyloid Core Protein May Account for Its Deposition and Stability in Alzheimer's Disease.” Journal of Biological Chemistry 268: 3072–3083. [PubMed] [Google Scholar]
  38. Roher, A. E. , Lowenson J. D., Clarke S., et al. 1993. “Beta‐Amyloid‐(1‐42) is a Major Component of Cerebrovascular Amyloid Deposits: Implications for the Pathology of Alzheimer Disease.” Proceedings of the National Academy of Sciences of the United States of America 90: 10836–10840. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Rostagno, A. , Cabrera E., Lashley T., and Ghiso J.. 2022. “N‐Terminally Truncated Aβ4‐x Proteoforms and Their Relevance for Alzheimer's Pathophysiology.” Translational Neurodegeneration 11: 30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Saido, T. C. , Iwatsubo T., Mann D. M. A., Shimada H., Ihara Y., and Kawashima S.. 1995. “Dominant and Differential Deposition of Distinct β‐Amyloid Peptide Species, AβN3(pE), in Senile Plaques.” Neuron 14: 457–466. [DOI] [PubMed] [Google Scholar]
  41. Scheres, S. H. W. , Ryskeldi‐Falcon B., and Goedert M.. 2023. “Molecular Pathology of Neurodegenerative Diseases by Cryo‐EM of Amyloids.” Nature 621: 701–710. [DOI] [PubMed] [Google Scholar]
  42. Schrempel, S. , Kottwitz A. K., Piechotta A., et al. 2024. “Identification of isoAsp7‐Aβ as a Major Aβ Variant in Alzheimer's Disease, Dementia With Lewy Bodies and Vascular Dementia.” Acta Neuropathologica 148: 78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Shin, Y. , Cho H., Fukumoto H., et al. 2003. “Aβ Species, Including IsoAsp23 Aβ, in Iowa‐Type Familial Cerebral Amyloid Angiopathy.” Acta Neuropathologica 105: 252–258. [DOI] [PubMed] [Google Scholar]
  44. Shinohara, M. , Murray M. E., Frank R. D., et al. 2016. “Impact of Sex and APOE4 on Cerebral Amyloid Angiopathy in Alzheimer's Disease.” Acta Neuropathologica 132: 225–234. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Sowade, R. F. , and Jahn T. R.. 2017. “Seed‐Induced Acceleration of Amyloid‐β Mediated Neurotoxicity In Vivo.” Nature Communications 8: 512. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Thal, D. R. , Rüb U., Orantes M., and Braak H.. 2002. “Phases of Aβ‐Deposition in the Human Brain and Its Relevance for the Development of AD.” Neurology 58: 1791–1800. [DOI] [PubMed] [Google Scholar]
  47. Tomidokoro, Y. , Rostagno A., Neubert T. A., et al. 2010. “Iowa Variant of Familial Alzheimer's Disease.” American Journal of Pathology 176: 1841–1854. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Tontsch, U. , and Bauer H. C.. 1989. “Isolation, Characterization, and Long‐Term Cultivation of Porcine and Murine Cerebral Capillary Endothelial Cells.” Microvascular Research 37: 148–161. [DOI] [PubMed] [Google Scholar]
  49. Van Dyck, C. H. , Swanson C. J., Aisen P., et al. 2023. “Lecanemab in Early Alzheimer's Disease.” New England Journal of Medicine 388: 9–21. [DOI] [PubMed] [Google Scholar]
  50. Vassar, R. , Bennett B. D., Babu‐Khan S., et al. 1999. “Beta‐Secretase Cleavage of Alzheimer's Amyloid Precursor Protein by the Transmembrane Aspartic Protease BACE.” Science 286: 735–741. [DOI] [PubMed] [Google Scholar]
  51. Vemuri, P. , Wiste H. J., Weigand S. D., et al. 2010. “Effect of Apolipoprotein E on Biomarkers of Amyloid Load and Neuronal Pathology in Alzheimer Disease.” Annals of Neurology 67: 308–316. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Walker, L. , Simpson H., Thomas A. J., and Attems J.. 2024. “Prevalence, Distribution, and Severity of Cerebral Amyloid Angiopathy Differ Between Lewy Body Diseases and Alzheimer's Disease.” Acta Neuropathologica Communications 12: 28. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Xia, Z. , Prescott E. E., Urbanek A., et al. 2024. “Co‐Aggregation With Apolipoprotein E Modulates the Function of Amyloid‐β in Alzheimer's Disease.” Nature Communications 15: 4695. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Yamada, M. 2000. “Cerebral Amyloid Angiopathy: An Overview.” Neuropathology 20: 8–22. [DOI] [PubMed] [Google Scholar]
  55. Yang, Y. , Arseni D., Zhang W., et al. 2022. “Cryo‐EM Structures of Amyloid‐β 42 Filaments From Human Brains.” Science 375: 167–172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Yang, Y. , Murzin A. G., Peak‐Chew S., et al. 2023. “Cryo‐EM Structures of Aβ40 Filaments From the Leptomeninges of Individuals With Alzheimer's Disease and Cerebral Amyloid Angiopathy.” Acta Neuropathologica Communications 11: 191. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Appendix S1: jnc70335‐sup‐0001‐AppendixS1.pdf.

JNC-169-0-s001.pdf (1.5MB, pdf)

Data Availability Statement

The quantitative data is available as Appendix S1 accompanying this article. Additional data that support the findings of this study are available from the corresponding author upon reasonable request. A preprint of this article was posted on BioRxiv on June 15, 2025 (doi: https://doi.org/10.1101/2025.06.11.658821).

Articles from Journal of Neurochemistry are provided here courtesy of Wiley

RESOURCES