Alpha-1-acid glycoprotein as a potential serum biomarker for cerebral amyloid angiopathy

Abstract

Background

Cerebral amyloid angiopathy (CAA) is a form of cerebral small vessel disease (SVD) associated with Alzheimer's disease, intracerebral hemorrhage, and cognitive decline. Despite its clinical significance, no reliable serum biomarker exists for early diagnosis or monitoring of disease progression.

Objective

This study hypothesizes that α1-acid glycoprotein (α1-AGP) and other serum biomarkers can aid CAA diagnosis and assessment using gel-based mass spectrometry. A comparative analysis was performed to investigate associations between serum biomarkers and radiological scores.

Methods

Serum proteins from individuals with probable or possible CAA (n = 10), classified using the modified Boston criteria, and age-matched controls (n = 10) were analyzed via two-dimensional differential gel electrophoresis (2D-DIGE) and matrix-assisted laser desorption/ionization time-of-flight tandem mass spectrometry (MALDI-TOF/TOF-MS). Candidate proteins were validated using enzyme-linked immunosorbent assay (ELISA). Outcome measures included biomarker diagnostic accuracy, assessed by receiver operating characteristic (ROC) curve analysis, and correlations between α1-AGP levels and CAA-SVD scores.

Results

Four proteins—hemopexin, complement C3, complement C9, and α1-AGP—were significantly elevated, while apolipoprotein A-1 was reduced in the CAA group. ELISA confirmed higher α1-AGP levels in individuals with CAA (p < 0.0001). ROC analysis demonstrated that α1-AGP could indicate the presence of CAA with a sensitivity and specificity of 1.00 (95%CI: 1.000, 1.000). Additionally, α1-AGP levels correlated with the CAA-SVD score (R² = 0.783).

Conclusions

α1-AGP may serve as a novel serum biomarker for CAA. Larger cohorts and external validation are required to substantiate these findings and determine their clinical relevance.

Keywords

α1-acid glycoproteins Alzheimer's disease biomarkers cerebral amyloid angiopathy cerebral small vessel diseases hemopexin microbleeds

Introduction

Cerebral amyloid angiopathy (CAA) is a form of cerebral small vessel disease (SVD) characterized by the progressive deposition of amyloid-β (Aβ) in the leptomeningeal and cortical arteries of the brain.^1,2 CAA is a leading cause of spontaneous intracerebral hemorrhage and contributes significantly to cognitive decline in older adults. The dysfunction of the vascular Aβ clearance system is considered the primary pathological mechanism in CAA. Although a definitive diagnosis requires autopsy-based brain tissue examination, clinical diagnosis is largely based on magnetic resonance imaging (MRI) findings, as outlined by the modified Boston criteria.³ However, the early identification of CAA—before MRI lesions are detectable—remains a significant challenge. An appealing approach for early CAA detection involves clinically accessible, minimally invasive tests. Although elevated serum complement C3 levels in individuals with CAA compared to those with mild cognitive impairment, and C3 has been proposed as a potential biomarker for CAA,^4,5 and reduced cerebrospinal fluid (CSF) concentrations of Aβ₄₀ and Aβ₄₂ have shown potential as diagnostic indicators,⁶ no reliable blood-based biomarkers for early CAA diagnosis or disease progression have been established.⁷ In contrast to brain biopsy and lumbar puncture, blood collection is minimally invasive, making it a promising tool for routine clinical practice.^8,9

This study aimed to analyze serum samples from CAA patients using the two-dimensional difference gel electrophoresis (2D-DIGE) technique, a well-established method for identifying diagnostic biomarkers.¹⁰ Following protein identification, expression levels were quantitatively assessed using enzyme-linked immunosorbent assay (ELISA), and we also examined the correlation between protein expression levels and SVD MRI scores to assess their clinical relevance. 2D-DIGE has been effectively used to examine protein expression profiles and is widely employed in cancer and neurodegenerative disease biomarker discovery.^10–12 Following protein identification, expression levels were quantitatively assessed using ELISA. Additionally, we examined the correlation between protein expression levels and SVD MRI scores.^13–16

Methods

Patients and clinical samples

We prospectively registered patients who visited the memory and neurology departments based on the following inclusion criteria: (1) probable or possible CAA as defined by the modified Boston criteria,¹⁷ (2) CAA was suspected due to the presence of numerous lobar cerebral microbleeds (CMBs) and cortical superficial siderosis (cSS) although the modified Boston criteria was not strictly met because of the presence of a few deep CMBs, (3) availability of blood samples and brain MRI, and (4) no history of major systemic diseases. The sample size (n = 10 per group) was determined as a preliminary investigation for future large-scale studies. Control participants were selected based on the absence of CAA-related MRI findings and no history of active systemic inflammatory diseases. Potential confounding factors, including underlying inflammatory or metabolic conditions that could influence biomarker levels, were considered. To mitigate these effects, individuals with known systemic inflammatory diseases were excluded. Ten patients with CAA, of which eight was probable CAA based on the modified Boston criteria,¹⁷ and two had a few deep microbleeds without hypertension but were included in the CAA group because CAA was suspected due to numerous lobar microbleeds, and ten controls without CAA were enrolled between April 2015 and March 2019. The control group included 2 patients with Parkinson's disease, 1 with motor neuron disease, 1 with migraine, 1 with dystonia, 1 with lacunar infarction, and 4 without neurological conditions.

Preparation of serum proteins

Serum samples were collected in EDTA-2Na tubes, centrifuged at 1000 × g for 15 min, and stored at −80°C. This study was approved by the ethics committee of Mie University (registration numbers: 2092 and H2018-032). Written informed consent was obtained from all participants or their families, and the study adhered to the principles of the Declaration of Helsinki. To reduce interference from high-abundance non-specific proteins, six proteins—albumin, IgG, α-1 antitrypsin, IgA, transferrin, and haptoglobin—were removed using the Multiple Affinity Removal Spin Cartridge Human-6HC (0.45 mL; Agilent Technologies, Santa Clara, CA, USA). Non-protein components, such as salts, lipids, and nucleic acids, were further removed using a 2D Clean-Up Kit (GE Healthcare, England, UK). The processed samples were resuspended in DIGE buffer (30 mM Tris-HCl, 7 M urea, 2 M thiourea, 4% (w/v) 3-((3-cholamidopropyl) dimethylammonio) propanesulfonate (CHAPS), and a protease inhibitor cocktail, pH 8.5). Protein concentrations were determined using the Bradford assay (Thermo Fisher Scientific, Waltham, MA, USA), with bovine serum albumin as the standard. All samples were then diluted in DIGE buffer to a final protein concentration of 2.5 μg/μL.

2D-DIGE

All samples were labelled with CyDyes, specifically developed for 2D-DIGE (GE Healthcare). Twenty-five micrograms of total protein per sample were labelled with 200 pmol of CyDyes and incubated on ice for 30 min in the dark. Five samples from control patients were initially labelled with Cy5, while five samples from patients with CAA were labelled with Cy3. To minimize the preferential binding of CyDye to a group-specific protein, the labelling was switched for the remaining samples: five control samples were labelled with Cy3, and the remaining five CAA patient samples were labelled with Cy5.

An internal standard was created by pooling equal amounts of all samples and labelling them with Cy2. The Cy2 labelled standard was run on all gels, enabling spot matching and normalization of signals across different gels. Equal amounts of Cy2-, Cy3-, and Cy5-labelled samples were mixed and combined with an equal volume of 2× sample buffer (7 M urea, 2 M thiourea, 4% (w/v) CHAPS, 130 mM dithiothreitol (DTT), 2% IPG buffer (pH 3–10; GE Healthcare), and a protease inhibitor cocktail). After incubation on ice for 10 min in the dark, the samples were combined with rehydration buffer (7 M urea, 2 M thiourea, 4% (w/v) CHAPS, 13 mM DTT, 1% IPG buffer (pI 3–10), and a protease inhibitor cocktail) and applied to IPG strips (pI 3–10 NL strips, 24 cm; GE Healthcare) for rehydration over 16 h.

Isoelectric focusing was performed using an Ettan IPGphor 3 (GE Healthcare). The IPG strips were treated to reduce and alkylate disulfide bonds with 10 mg/mL DTT and 25 mg/mL iodoacetamide, respectively. Standard sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was subsequently performed using ten vertical 12.5% gels on the Ettan DALTsix Large Format Vertical System (GE Healthcare). Gel images from ten patients with CAA and ten controls were obtained using a Typhoon FLA 9500 scanner (GE Healthcare). Intragel spot detection and intergel matching were performed using the differential in-gel analysis and biological variation analysis modules in DeCyder 2D software version 7.2 (GE Healthcare). Spot integration was carried out based on parameters including spot maximum slope, spot area, spot volume, and spot peak height. Protein expression values were statistically analyzed using Student's t-test within the DeCyder 2D software.

Protein identification by in-gel digestion and MALDI-TOF/TOF ms

The protein identification method was adapted from Kondo et al. with slight modifications.¹⁸ Coomassie brilliant blue (CBB)-stained sections of the differential protein spots in the two-dimensional electrophoresis (2DE) gel were excised, decolorized, dehydrated, and incubated with trypsin solution (Promega Corporation, Madison, WI, USA) for overnight digestion at 37°C.

Tryptic peptides were extracted using 45% acetonitrile/0.1% trifluoroacetic acid (TFA), concentrated, and mixed with a saturated α-cyanohydroxycinnamic acid (Wako Pure Chemical, Osaka, Japan) matrix solution in a 1:0.5 ratio. The mixture was spotted on a stainless-steel target plate. Mass analysis was performed using matrix-assisted laser desorption ionization time-of-flight tandem mass spectrometry (MALDI-TOF/TOF MS; 4800 Plus MALDI-TOF/TOF^TM Analyzer; AB SCIEX, Framingham, MA, USA). Protein identification was performed using the MS/MS ion search tool in ProteinPilot software (AB SCIEX).

Quantitative evaluations

We quantitatively measured the protein expression levels using ELISA to assess whether the serum levels of each protein were altered in patients with CAA compared to controls. Specific ELISA kits for each target protein were used: Human Hemopexin SimpleStep ELISA Kit (Abcam, ab221838), Human Complement C3 ELISA Kit (Abcam, ab108823), Human Complement C9 SimpleStep ELISA Kit (Abcam, ab287178), Human α1-Acid Glycoprotein SimpleStep ELISA Kit (Abcam, ab243675), and Human Apolipoprotein A1 ELISA Kit (Proteintech, KE00157). Optical density was measured at 450 nm using a microplate reader, and the standard curve was constructed with known antibody concentrations. Sample concentrations were interpolated from this standard curve.

SVD scores

Brain MRI was conducted on all patients using a 3 T MR scanner with either an 8- or 32-channel phased-array head coil. Diagnostic imaging included 3D-T1-weighted imaging, T2-weighted imaging, 3D-fluid-attenuated inversion recovery (FLAIR), 3D-double inversion recovery (DIR), and susceptibility-weighted imaging (SWI). The total SVD score was calculated by awarding 1 point for the severity of each of the four markers: lacunar infarcts, CMBs, basal ganglia-perivascular spaces (PVSs), and white matter hyperintensities (WMHs). The score ranged from 0 to 4.^14,15 The CAA-SVD score was the sum of points for the severity of four markers: lobar cerebral microbleeds (Figure 1A), cSS (Figure 1A), centrum semiovale perivascular spaces (CSO-PVS, Figure 1B), WMH (Figure 1C), and, with a minimum score of 0 and a maximum of 6.¹³ The modified CAA-SVD score incorporated additional markers, including posterior dominant WMH (Figure 1C) and cortical microinfarcts (CMIs, Figure 1D) associated with CAA, yielding a score range from 0 to 8.¹⁶ CMBs were evaluated based on the Microbleed Anatomical Rating Scale (MARS)¹⁹ on SWI. All scores were independently assessed by two neurologists (H.I. and H.M.) to ensure accuracy and reduce bias. Intraclass correlation coefficients for the total SVD score, CAA-SVD score, and modified CAA-SVD score were 0.934, 0.982, and 0.979, respectively.

Figure 1.

Representative MRI findings of CAA. The arrows indicate lobar cerebral microbleeds (arrows, A), with cortical superficial siderosis visible on the SWI sequences (arrowheads, A). Enlarged perivascular spaces in the centrum semiovale are evident on T2-weighted imaging (arrows, B). Posteriorly distributed white matter hyperintensities are observed on FLAIR imaging (C). A microinfarct localized within the cortex is highlighted on the 3D-DIR sequence (arrow, D). CAA: cerebral amyloid angiopathy; FLAIR: fluid-attenuated inversion recovery; SWI: susceptibility-weighted imaging; 3D-DIR: three-dimensional double inversion recovery.

Statistical analysis

Descriptive statistical analyses were performed using SPSS Statistics version 22 (IBM Corp.). Continuous variables were summarized as medians with interquartile ranges (IQRs). Categorical variables were analyzed using chi-squared or Fisher's exact tests, while continuous variables were compared using either the Student's t-test or the Mann–Whitney U test. Linear regression analysis was conducted to assess the association between the SVD scores and the serum concentrations of each protein. Receiver operating characteristic (ROC) curves were constructed to evaluate the sensitivity and specificity of each protein in distinguishing CAA from non-CAA cases. A significance threshold of p < 0.05 was applied for all analyses.

Results

Patient characteristics

The clinical characteristics of patients in the CAA and non-CAA groups are summarized in Table 1. No significant differences were observed between the groups with respect to age, sex, social history, medical history, or medication use. Additionally, no acute illnesses were noted during treatment. MRI findings revealed no significant differences except for those specified by the modified Boston criteria. However, significant differences in SVD markers were observed between the CAA and control groups, with the CAA group demonstrating higher SVD scores.

Table 1.

Characteristics of the patients with CAA and control.

	CAA	Control	p
	n = 10	n = 10	p
Demographics
Mean age (y)	73.6 ± 6.9	77.4 ± 7.4	0.837
Male sex, (n, %)	7 (70.0%)	5 (50.0%)	0.325
MMSE score	20.5 ± 5.5	29.1 ± 1.4	0.006*
Hypertension (n, %)	7 (70.0%)	2 (20.0%)	0.035*
Dyslipidemia (n, %)	3 (30.0%)	0 (0.0%)	0.060
Diabetes mellitus (n, %)	2 (20.0%)	0 (0.0%)	0.136
Malignant tumor (n, %)	2 (20.0%)	0 (0.0%)	0.136
Total SVD score, median (IQR)
Lacune (0-1)	1 (0-1)	0 (0-0)	0.001*
CMBs (0-1)	1 (1-1)	0 (0-0)	0.001*
Basal ganglia PVS (0-1)	1 (0-1)	0.5 (0-1)	0.054
WMH (0-1)	1 (1-1)	1 (0-1)	0.331
Total (0-4)	4 (3-4)	2 (1-3)	0.001*
CAA-SVD score, median (IQR)
Lobar CMBs (0-2)	2 (2-2)	0 (0-0)	<0.001*
cSS (0-2)	2 (1-2)	0 (0-0)	0.001*
Centrum semiovale PVS (0-1)	1 (0-1)	0 (0-1)	0.081
WMH (0-1)	1 (1-1)	1 (0-1)	0.331
Total (0-6)	5 (4-6)	1 (0-2)	0.001*
Modified CAA-SVD score, median (IQR)
Posterior distribution of WMH (0-1)	1 (0-1)	0 (0-1)	0.004*
CMIs due to CAA (0-1)	0.5 (0-1)	0 (0-0)	0.054
Total (0-8)	6.5 (5-7)	1.5 (1-2)	0.001*

2D-DIGE

To investigate serum protein profiles in CAA patients and controls, 2D-DIGE and MALDI-TOF/TOF/MS analyses were performed.^18,20 Figure 2A presents a representative image of a 2D-DIGE gel containing two samples: one labelled with Cy5 (CAA) and the other with Cy3 (control). Green spots on the gel represent downregulated serum proteins in CAA patients compared to controls, while red spots represent upregulated proteins. Yellow spots indicate proteins with similar expression levels in both groups.

Figure 2.

Images of serum proteins detected by 2D-DIGE (A) and CBB staining (B). (A) 2D-DIGE images of the CAA (Cy5) and control (Cy3) groups. Serum proteins (25 μg) were labelled with Cy3 (control) and Cy5 (CAA) dyes, mixed, and subjected to 2D-DIGE analysis. Green spots represent proteins that are downregulated in CAA relative to the control group, while red spots correspond to upregulated proteins. Yellow spots indicate proteins with unchanged expression levels. (B) Typical CBB-stained 2DE map of serum samples. Coomassie Brilliant Blue (CBB) staining of the 2DE map. A typical CBB-stained two-dimensional electrophoresis (2DE) map displaying the serum protein profile, emphasizing differentially expressed proteins between the CAA and control groups.

Comparative analysis revealed significant differences in the expression levels of 51 protein spots between the CAA and control groups. These 51 spots were excised, digested in-gel, and analyzed via MALDI-TOF/TOF/MS. Of these, 28 spots failed to yield detectable or reproducible results due to low protein concentrations. Additionally, five spots identified as haptoglobin were excluded as this protein had been removed during sample preparation using the Multiple Affinity Removal Spin Cartridge Human-6HC. Ultimately, 18 spots were selected as containing a single differentially expressed protein. Among these, five spots could be identified by mass spectrometry analysis followed by database matching (Figure 2B). The spots with significantly upregulated proteins were identified as hemopexin, complement C3, complement C9, and α1-acid glycoprotein (α1-AGP), while one protein, apolipoprotein A-1, was significantly downregulated in the CAA group.

Assessment using ELISA

The expression levels of five proteins (Hemopexin, Complement C9, Complement C3, α1-AGP, and Apolipoprotein A-1) were quantitatively measured using ELISA. Figure 3 displays the expression levels of each protein in the CAA and control groups. While no statistically significant differences were found between the two groups for Hemopexin, Complement C9, and Apolipoprotein A-1 (p = 0.364, 0.226, and 0.09, respectively), significant differences were observed for Complement C3 (p = 0.023) and α1-AGP (p < 0.0001), which were elevated in the CAA group. Figure 4 illustrates the ROC curve for each protein, with the area under the curve (AUC) for α1-AGP demonstrating significant discriminatory ability between the CAA and control groups. Receiver-operating curve of each protein. The AUC for each protein was obtained from all patients, illustrating the diagnostic potential of the respective biomarkers in distinguishing CAA from control. Serum α1-AGP levels distinguished CAA from controls at a cut-off value of 268.5 mg/dL, with a sensitivity and specificity of 1.00 (95% CI: 1.000, 1.000). Complement C3 exhibited a sensitivity of 0.70 and a specificity of 0.20 (95% CI: 0.600, 1.000) at a cut-off value of 307.2 mg/dL.

Figure 3.

Comparison of serum proteins between CAA and control. Serum levels of each biomarker in the CAA and control groups were measured using ELISA. The levels of Complement C3 and α1-AGP showed significant differences between the two groups (**p < 0.05, ***p < 0.0001), while no statistically significant differences were observed for the other biomarkers unless indicated by **p < 0.05 or ***p < 0.0001.

Figure 4.

Receiver operating characteristic curve of each protein. The area under the curve (AUC) for each protein was obtained from all patients, illustrating the diagnostic potential of the respective biomarkers in distinguishing CAA from control.

Correlation between protein expression levels and SVD scores on MRI

The relationship between SVD scores on MRI and the concentration of each protein was analyzed (Figure 5 and Supplemental Figure 1). No significant correlations were observed between the concentrations of C3, Hemopexin, C9, or Apolipoprotein A-1 and SVD scores. However, a significant correlation was identified between α1-AGP and both the CAA-SVD score (R² = 0.783) and the modified CAA-SVD score (R² = 0.652). Linear regression analysis demonstrated a strong relationship between α1-AGP concentration and CAA-SVD scores, with the coefficient of determination for CAA-SVD being R² = 0.783. Raw data for the 20 participants, including clinical and biomarker data, are provided in the Supplemental Material (Supplemental Table 1).

Figure 5.

Linear regression models of the associations between each MRI SVD-score and the concentration of complement C3 and α1-AGP levels. CAA: cerebral amyloid angiopathy; SVD: small vessel disease.

Discussion

The major findings of the present study are as follows. First, the 2D-DIGE analysis revealed potential biomarkers of CAA, specifically Hemopexin, C3, C9, α1-AGP, and Apolipoprotein A-1. Second, subsequent ELISA assays identified α1-AGP as a novel serum biomarker for CAA. Third, ROC curve analysis demonstrated that serum α1-AGP levels exceeding 268.5 mg/dL could indicate CAA. Fourth, the serum α1-AGP level was strongly correlated with the CAA-SVD score.

This study provides new insights into identifying potential biomarkers of CAA by combining mass spectrometry with a gel-based method for candidate detection, immunoassay-based validation techniques, and comparative radiological biomarker score analysis to confirm clinical utility.

Dementia represents a significant global social and medical challenge.²¹ Alzheimer's disease (AD), the most prevalent form of dementia, is characterized by the deposition of Aβ plaques and phosphorylated tau protein in the brain.⁷ CAA is neuropathologically detected in approximately 80% of AD cases.²² Moreover, both ischemic and hemorrhagic lesions contribute to the progression of vascular dementia as well as AD.^23,24 The discovery phase of this study, employing 2D-DIGE, identified a list of proteins differentially expressed between the serum samples of CAA patients and controls, based on a semiquantitative estimation of protein relative abundances. However, as some of these may represent false positives, these proteins should be considered candidate biomarkers until further validated.²⁵ Developing an independent analytical approach to validate potential markers is a crucial step in biomarker research. ELISA, a widely used immunoassay technique, is preferred for its high sensitivity and throughput, making it suitable for clinical diagnostic development.⁷ Following a mass spectrometry-based biomarker discovery workflow, α1-AGP and complement C3 were identified as potential biomarkers in serum samples. Notably, complement C3 has previously been identified as a potential serum diagnostic biomarker for CAA.⁴ However, its clinical utility remains contentious, as elevated levels of complement C3 are also observed in other conditions, such as stroke and non-alcoholic fatty liver disease.^26,27 α1-AGP, also known as orosomucoid, has recently emerged as a novel mediator of astrocyte-microglia interactions in the CNS.²⁸ This glycoprotein is predominantly synthesized in the liver and plays a crucial role in modulating immune responses and maintaining inflammatory balance.²⁹ Elevated serum levels of α1-AGP are associated with a range of pathological conditions, including inflammation, infection, and cancer.^30,31 Recent studies have shown that α1-AGP levels measured 72 h after subarachnoid hemorrhage can predict delayed cerebral ischemia and unfavorable outcomes.²⁹ Beyond its role as an acute-phase protein, α1-AGP has also been implicated in chronic cerebral circulation insufficiency and diabetes mellitus in older populations, compared to healthy controls.³² Although the etiology of chronic cerebral circulation insufficiency in previous reports was unclear, it may have included cases of CAA.

In this study, the ROC analysis demonstrated that serum α1-AGP shows a sensitivity and specificity of 1.00 at a cut-off value of 268.5 mg/dL. These findings suggest that α1-AGP is a superior marker for distinguishing CAA in our cohort. However, further validation studies are needed to establish the clinical utility of α1-AGP, particularly given the potential for elevated levels in other pathological conditions.

We also found that the serum level of α1-AGP is associated with the CAA-SVD score on MRI. The CAA-SVD score has been established as a predictor of recurrent intracranial cerebral hemorrhage and as a utility marker for assessing the severity of CAA.³³ This score was developed to clinically assess CAA severity on MRI based on pathological validation using a neuropathologically defined CAA cohort. The higher the score, the more severe the CAA-related vasculopathic changes.¹³ A recent study has demonstrated that decreased CSF Aβ₄₀ and Aβ₄₂ levels are the earliest biomarkers in hereditary CAA, and these markers correlate with MRI biomarkers of both non-hemorrhagic and hemorrhagic changes.³⁴ However, no blood biomarker is currently available for the diagnosis of CAA. Our findings suggest that the serum level of α1-AGP might correlate with the progression of CAA, as both the CAA-SVD and modified CAA-SVD scores showed significant correlations with serum α1-AGP levels, whereas no such correlation was observed with the total SVD score.

Monoclonal antibodies targeting Aβ have recently received approval as disease-modifying treatments for AD, which have been shown to slow the progression of AD. However, a complication known as amyloid-related imaging abnormalities (ARIA) can occur during anti-Aβ monoclonal antibody treatment. CAA has been identified as a significant risk factor for the development of ARIA.^35,36 Consequently, assessing ARIA risk factors is essential for the safe administration of anti-amyloid therapies, as ARIA can result in severe neurological dysfunction or even death.³⁷ Therefore, a serum biomarker for CAA is crucial for selecting appropriate patients for anti-amyloid therapies.³⁴ The discovery of biomarkers has been an area of extensive research since the advent of proteomics.³⁸ If α1-AGP proves to be a reliable biomarker for CAA, it could facilitate safer administration of disease-modifying treatments for AD, such as aducanumab, lecanemab, and donanemab.

Although our study identified novel potential biomarkers for CAA, it has some limitations. First, the limited sample size may have influenced our findings. In our cohort, patients without inflammatory diseases, infections, acute stroke, or cancer were included. As such, our results cannot be generalized to individuals with these conditions. Further investigations involving larger cohorts, encompassing diverse clinical conditions, and external validation in independent cohorts are required to substantiate and extend these findings. Second, we did not evaluate the longitudinal progression of biomarkers. The significant correlation observed between serum α1-AGP levels and CAA radiological markers in our study suggests that α1-AGP may serve as an indicator of CAA pathology progression. However, as our findings are based on a cross-sectional analysis, longitudinal studies are essential to elucidate biomarker dynamics over time. Additionally, a longitudinal investigation tracking temporal changes in α1-AGP levels alongside CAA radiological markers, such as WMH, PVS, CMBs, and cSS, would provide further validation. Third, 2 patients with a small number of deep microbleeds but no history of hypertension were included in the CAA group. These 2 cases in the CAA group were classified as ‘CAA’ because of the presence of characteristic CAA MRI markers including numerous lobar CMBs and cSS. To address this, we performed ELISA and ROC analyses excluding these cases, which α1-AGP yielded consistent results (Supplemental Figures 2 and 3). However, a recent neuropathological study has showed deep microbleeds on MRI in 15% of CAA patients, suggesting that CAA may not be ruled out even in the presence of deep microbleeds.³⁹ The results of these two cases suggest that α1-AGP may be a potential biomarker for the presence of CAA even in cases with mixed CMBs having numerous lobar CMBs and/or cSS. Fourth, the control group included individuals with various neurological conditions, which may introduce potential confounding factors. Finally, we did not include CSF biomarker data in this study. Previous research has shown that amyloid biomarkers in CSF are the earliest indicators of CAA, preceding MRI biomarkers.⁴⁰ These findings suggest that serum α1-AGP elevation may follow earlier changes in CSF biomarkers, highlighting the need for future studies that incorporate CSF data to further validate our results.

Using 2D-DIGE, ELISA, and comparative analysis of MRI CAA-SVD scores, we identified and validated novel serum biomarkers. In conclusion, α1-AGP appears to be a promising biomarker for diagnosing and monitoring CAA progression. Future research should focus on longitudinal studies to evaluate biomarker stability over time and validate these findings in diverse populations with varying CAA severities. Additionally, integrating α1-AGP with other biomarkers may enhance diagnostic accuracy and clinical applicability.

Supplemental Material

sj-docx-1-alz-10.1177_13872877251333802 - Supplemental material for Alpha-1-acid glycoprotein as a potential serum biomarker for cerebral amyloid angiopathy

Supplemental material, sj-docx-1-alz-10.1177_13872877251333802 for Alpha-1-acid glycoprotein as a potential serum biomarker for cerebral amyloid angiopathy by Akisato Nishigaki, Hidehiro Ishikawa, Yamato Nishiguchi, Kei Tachibana, Natsuko Kato, Kana Matsuda, Yurie Mori, Hirofumi Matsuyama, Keita Matsuura, Yuichiro Ii, Hideaki Wakita, Shinji Oikawa, Hidekazu Tomimoto and Akihiro Shindo in Journal of Alzheimer's Disease

Supplemental Material

sj-xlsx-2-alz-10.1177_13872877251333802 - Supplemental material for Alpha-1-acid glycoprotein as a potential serum biomarker for cerebral amyloid angiopathy

Supplemental material, sj-xlsx-2-alz-10.1177_13872877251333802 for Alpha-1-acid glycoprotein as a potential serum biomarker for cerebral amyloid angiopathy by Akisato Nishigaki, Hidehiro Ishikawa, Yamato Nishiguchi, Kei Tachibana, Natsuko Kato, Kana Matsuda, Yurie Mori, Hirofumi Matsuyama, Keita Matsuura, Yuichiro Ii, Hideaki Wakita, Shinji Oikawa, Hidekazu Tomimoto and Akihiro Shindo in Journal of Alzheimer's Disease

Footnotes

Acknowledgments

We thank Ms. Kaori Kawamura and Ms. Futaba Ikeda for their technical assistance.

ORCID iD

Akihiro Shindo

Ethical considerations

The study was approved by the Ethical Committees of Mie University (Permit Numbers 2092 and H2018-032).

Consent to participate

Written informed consent was obtained from all participants or their families.

Author contributions

Akisato Nishigaki: Data curation; Formal analysis; Validation; Writing - original draft; Hidehiro Ishikawa: Data curation; Formal analysis; Methodology; Writing - review & editing; Yamato Nishiguchi: Data curation; Formal analysis; Kei Tachibana: Data curation; Formal analysis; Natsuko Kato: Data curation; Formal analysis; Kana Matsuda: Data curation; Investigation; Yurie Mori: Formal analysis; Investigation; Hirofumi Matsuyama: Data curation; Investigation; Keita Matsuura: Investigation; Yuichiro Ii: Investigation; Methodology; Writing - review & editing; Hideaki Wakita: Methodology; Writing - review & editing; Shinji Oikawa: Formal analysis; Methodology; Validation; Writing - review & editing; Hidekazu Tomimoto: Methodology; Supervision; Writing - review & editing; Akihiro Shindo: Conceptualization; Funding acquisition; Methodology; Project administration; Writing - review & editing

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the KAKENHI Grant-in-Aid for Scientific Research C (Grant Number: 21K07433) and Research B (Grant Numbers: 23K27838 and 24K02366).

Japan Society for the Promotion of Science, (grant number 21K07433, 23K27838, 24K02366).

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability

The raw data supporting the findings of this study, including ELISA assay results and MRI SVD score, are available in the Supplemental Table 1.

Supplemental material

Supplemental material for this article is available online.

References

Charidimou

Pantoni

Love

. The concept of sporadic cerebral small vessel disease: a road map on key definitions and current concepts. Int J Stroke 2016; 11: 6–18.

Pantoni

. Cerebral small vessel disease: from pathogenesis and clinical characteristics to therapeutic challenges. Lancet Neurol 2010; 9: 689–701.

Charidimou

Boulouis

. Clinical diagnosis of probable cerebral amyloid angiopathy: diagnostic accuracy meta-analysis of the Boston Criteria. Stroke 2022; 53: 3679–3687.

Saito

Yamashiro

Yamauchi

, et al. Complement 3 is a potential biomarker for cerebral amyloid angiopathy. J Alzheimers Dis 2022; 89: 381–387.

Schroeder

Robertson

Hughes

TAT

. Cerebral amyloid angiopathy: an update. J Neurol 2023; 270: 2809–2811.

van Etten

Verbeek

van der Grond

, et al. β-Amyloid in CSF. Neurology 2017; 88: 169–176.

Lista

Faltraco

Prvulovic

, et al. Blood and plasma-based proteomic biomarker research in Alzheimer's disease. Prog Neurobiol 2013; 101-102: 1–17.

Puchades

Hansson

Nilsson

, et al. Proteomic studies of potential cerebrospinal fluid protein markers for Alzheimer's disease. Mol Brain Res 2003; 118: 140–146.

Yang

M-H

Yang

Y-H

C-Y

, et al. Activity-dependent neuroprotector homeobox protein: a candidate protein identified in serum as diagnostic biomarker for Alzheimer's disease. J Proteomics 2012; 75: 3617–3629.

10.

Kitamura

Usami

Ichihara

, et al. Plasma protein profiling for potential biomarkers in the early diagnosis of Alzheimer’s disease. Neurol Res 2017; 39: 231–238.

11.

Tajima

Kito

Yoshida

, et al. Calreticulin as a novel potential metastasis-associated protein in myxoid liposarcoma, as revealed by two-dimensional difference gel electrophoresis. Proteomes 2019; 7: 13.

12.

Wang

Sun

Guo

, et al. Tetranectin and apolipoprotein A-I in cerebrospinal fluid as potential biomarkers for Parkinson's disease. Acta Neurol Scand 2010; 122: 350–359.

13.

Charidimou

Martinez-Ramirez

Reijmer

, et al. Total magnetic resonance imaging burden of small vessel disease in cerebral amyloid angiopathy: an imaging-pathologic study of concept validation. JAMA Neurol 2016; 73: 994–1001.

14.

Staals

Makin

Doubal

, et al. Stroke subtype, vascular risk factors, and total MRI brain small-vessel disease burden. Neurology 2014; 83: 1228–1234.

15.

Klarenbeek

van Oostenbrugge

Rouhl

, et al. Ambulatory blood pressure in patients with lacunar stroke: association with total MRI burden of cerebral small vessel disease. Stroke 2013; 44: 2995–2999.

16.

Matsuda

Shindo

, et al. Investigation of hypertensive arteriopathy-related and cerebral amyloid angiopathy-related small vessel disease scores in patients from a memory clinic: a prospective single-centre study. BMJ Open 2021; 11: e042550.

17.

Greenberg

Charidimou

. Diagnosis of cerebral amyloid angiopathy: evolution of the Boston Criteria. Stroke 2018; 49: 491–497.

18.

Kondo

Hirohashi

. Application of highly sensitive fluorescent dyes (CyDye DIGE Fluor saturation dyes) to laser microdissection and two-dimensional difference gel electrophoresis (2D-DIGE) for cancer proteomics. Nat Protoc 2006; 1: 2940–2956.

19.

Gregoire

Chaudhary

Brown

, et al. The Microbleed Anatomical Rating Scale (MARS): reliability of a tool to map brain microbleeds. Neurology 2009; 73: 1759–1766.

20.

Oikawa

Kobayashi

Kitamura

, et al. Proteomic analysis of carbonylated proteins in the monkey substantia nigra after ischemia-reperfusion. Free Radic Res 2014; 48: 694–705.

21.

Walsh

Wallace

Kuhn

, et al. Population-level interventions for the primary prevention of dementia: a complex evidence review. EClinicalMedicine 2024; 70: 102538.

22.

Thal

Griffin

de Vos

, et al. Cerebral amyloid angiopathy and its relationship to Alzheimer's disease. Acta Neuropathol 2008; 115: 599–609.

23.

Greenberg

Bacskai

Hernandez-Guillamon

, et al. Cerebral amyloid angiopathy and Alzheimer disease - one peptide, two pathways. Nat Rev Neurol 2020; 16: 30–42.

24.

Sweeney

Kisler

Montagne

, et al. The role of brain vasculature in neurodegenerative disorders. Nat Neurosci 2018; 21: 1318–1331.

25.

Rifai

Gillette

Carr

. Protein biomarker discovery and validation: the long and uncertain path to clinical utility. Nat Biotechnol 2006; 24: 971–983.

26.

Han

Zhang

. Complement component C3: a novel biomarker participating in the pathogenesis of non-alcoholic fatty liver disease. Front Med (Lausanne) 2021; 8: 653293.

27.

Luo

Sun

, et al. Effects of adjuvant berberine therapy on acute ischemic stroke: a meta-analysis. Phytother Res 2023; 37: 3820–3838.

28.

Kim

Song

, et al. Astrocytic orosomucoid-2 modulates microglial activation and neuroinflammation. J Neurosci 2017; 37: 2878–2894.

29.

Zavori

Varnai

Molnar

, et al. Acute phase protein orosomucoid (alpha-1-acid glycoprotein) predicts delayed cerebral ischemia and 3-month unfavorable outcome after aneurysmal subarachnoid hemorrhage. Int J Mol Sci 2023; 24: 15267.

30.

Smith

Waters

. Pharmacokinetic and pharmacodynamic considerations for drugs binding to alpha-1-acid glycoprotein. Pharm Res 2018; 36: 30.

31.

Yue

Dai

, et al. Orosomucoid 1 promotes colorectal cancer progression and liver metastasis by affecting PI3K/AKT pathway and inducing macrophage M2 polarization. Sci Rep 2023; 13: 14092.

32.

Goncharov

Popova

Kudryavtsev

, et al. Immunological profile and markers of endothelial dysfunction in elderly patients with cognitive impairments. Int J Mol Sci 2024; 25: 1888.

33.

Boulouis

Charidimou

Pasi

, et al. Hemorrhage recurrence risk factors in cerebral amyloid angiopathy: comparative analysis of the overall small vessel disease severity score versus individual neuroimaging markers. J Neurol Sci 2017; 380: 64–67.

34.

Sembill

Lusse

Linnerbauer

, et al. Cerebrospinal fluid biomarkers for cerebral amyloid angiopathy. Brain Commun 2023; 5: fcad159.

35.

Aldea

Grimm

Gieschke

, et al. In silico exploration of amyloid-related imaging abnormalities in the gantenerumab open-label extension trials using a semi-mechanistic model. Alzheimers Dementia (N Y) 2022; 8: e12306.

36.

Hampel

Elhage

Cho

, et al. Amyloid-related imaging abnormalities (ARIA) and their radiological, biological and clinical characteristics: a plain language summary. Neurodegener Dis Manag 2024; 14: 51–62.

37.

Cummings

Apostolova

Rabinovici

, et al. Lecanemab: appropriate use recommendations. J Prev Alzheimers Dis 2023; 10: 362–377.

38.

Petricoin

Ardekani

Hitt

, et al. Use of proteomic patterns in serum to identify ovarian cancer. Lancet 2002; 359: 572–577.

39.

Huhndorf

Röcken

Flüh

, et al. Frequency of deep-seated cerebral microbleeds in patients with lobar hemorrhages and histopathological evidence for cerebral amyloid angiopathy. Front Neurol 2023; 14: 1146737.

40.

Koemans

Chhatwal

van Veluw

, et al. Progression of cerebral amyloid angiopathy: a pathophysiological framework. Lancet Neurol 2023; 22: 632–642.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.50 MB

0.01 MB