Genetic Validation of ABI3 p.Ser209Phe Variant and Its effects On Early Brain Pathology in Asymptomatic Elderly Individuals

doi:10.21203/rs.3.rs-8113148/v1

Download PDF

Research Article

Genetic Validation of ABI3 p.Ser209Phe Variant and Its effects On Early Brain Pathology in Asymptomatic Elderly Individuals

https://doi.org/10.21203/rs.3.rs-8113148/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Alzheimer’s disease (AD) has a strong genetic component, with APOE ε4 being the most established risk factor through its effects on beta-amyloid (Aβ) metabolism and microglial function. Recent genetic studies have also implicated microglial genes, such as the ABI3^S209F variant, to increased AD risk. As APOE ε4 and ABI3^S209F influence microglial pathways through distinct mechanisms, their combined analysis may provide novel insights into AD pathophysiology. Therefore, we investigated ABI3^S209F in the Finnish FinnGen cohort and in an imaging study of cognitively healthy older adults

Methods

We used FinnGen R12 data (> 500,000 individuals), including 8,490 ABI3^S209F carriers and 511,670 non-carriers, with survival analyses matched by sex and birth year. Disease endpoints (AD, dementia, neurodegenerative disorder) were defined from national health registries using harmonized ICD codes, medication, and reimbursement records. For the imaging study, 58 participants aged ≥ 50 years were recruited into three genotype-based groups (ABI3^S209F/APOE ε4, ABI3^S209F/APOE ε3, non-carriers). All imaging participants underwent structural MRI, [¹¹C]PiB PET for amyloid beta, [¹¹C]PK11195 PET for microglial activity, and a comprehensive neuropsychological battery.

Results

ABI3^S209F was significantly associated with increased risk of AD (OR = 1.22, p = 0.0012) and neurodegenerative disorders (OR = 1.21, p = 0.00023), but not with dementia (OR = 1.10, p = 0.06). Survival analyses indicated that ABI3^S209F carriers developed AD at an earlier age than non-carriers with the same APOE genotype. The carriers of ABI3^S209F and APOE ε4 had higher brain Aβ burden when compared to the ABI3^S209F carriers without APOE ε4 (SUVR 2.0 (0.7) vs. 1.67 (0.5); SUVR 2.0 (0.7) vs. 1.67 (0.5); mean (sd), p = 0.017). ABI3^S209F was not associated with global neuroinflammation, although subtle regional increases in [¹¹C]PK11195 binding were observed in ABI3^S209F ε4 carriers. No differences were found in brain volumes or cognition.

Conclusions

ABI3^S209F increases AD risk and is associated with earlier disease onset. The variant alone does not significantly influence cortical Aβ deposition, neuroinflammation, or brain structure. Its effect may be pronounced in combination with APOEε4.

Alzheimer’s disease

ABI3S209F

APOE ε4

Microglia

β-amyloid-

PET imaging

FinnGen

Alzheimer's disease (AD) has a strong genetic component, with the Apolipoprotein E ε4 (APOEε4) being the most common and well-established genetic risk factor (Corder et al., 1993; Guerreiro, Gustafson, & Hardy, 2012). APOEε4 is associated with a higher burden of beta-amyloid (Aβ) plaques in the brain in a dose-dependent manner (Castellano et al., 2011; Reiman et al., 2009), and it is thought to impair Aβ clearance (Verghese et al., 2013).

In recent years, genome-wide association studies (GWAS) have identified several additional risk genes for AD (C. Bellenguez et al., 2022; Jansen et al., 2019; D. P. Wightman et al., 2021), including ones with strong expression in microglia, such as ABI3. The ABI3 p.Ser209Phe variant (ABI3^S209F) was first linked to increased AD risk in 2017 with an odds ratio (OR) of 1.43 (Sims et al., 2017), and this association has been replicated in multiple cohorts (Conway et al., 2018; Dalmasso et al., 2019; Olive et al., 2020; Douglas P. Wightman et al., 2021).

ABI3 is highly expressed in microglial cells, with limited expression in neurons and other glial cell types (Conway et al., 2018; Satoh et al., 2017). Microglia have dual roles in AD pathology, initially protective through Aβ clearance, but later potentially harmful due to sustained inflammation (Z. Fan, Brooks, Okello, & Edison, 2017; Leng & Edison, 2021). ABI3 is upregulated in the cortex of AD patients and in Aβ mouse models (Castillo et al., 2017), and is thought to modulate microglial responses through interferon signalling (Sims et al., 2017), and actin cytoskeletal reorganization via the WAVE2 complex (Davidson, Ura, Thomason, Kalna, & Insall, 2013; Sekino et al., 2015). However, findings from murine Abi3 knockout models have been inconsistent as some show increased Aβ deposition (Karahan et al., 2021; Karahan et al., 2023), while others note a transient reduction (Ibanez et al., 2022). Despite the replicated association between ABI3^S209F and AD, evidence for its impact on in vivo AD biomarkers is limited. Notably, Olive et al. (2020) found no significant association between the variant and CSF Aβ42 or Aβ PET signal in cognitively normal individuals.

APOEε4 does not only influence Aβ accumulation but also directly modulates microglial function (Krasemann et al., 2017; Rodriguez, Tai, Ladu, & Rebeck, 2014; Ulrich et al., 2018). The APOEε4 isoform has been shown to alter microglial responses by limiting their ability to adopt a neuroprotective phenotype, impairing Aβ clearance and promoting their pro-inflammatory state (Liu et al., 2023; Nguyen et al., 2020; Yin et al., 2023). The dual role of APOEε4 in Aβ pathology and immune regulation makes it a particularly relevant modifier when exploring microglial activity and gene–gene interactions in AD.

Given that both APOEε4 and ABI3 influence microglial function, albeit through distinct molecular pathways, their combined analysis may provide a more comprehensive understanding of microglial contributions to AD pathophysiology. Investigating these genes in parallel allows for the exploration of potential gene–gene interactions, additive effects, or pathway convergence that may influence cortical Aβ deposition and neuroinflammation. Also, due to the genetic and pathological heterogeneity of AD, replication of risk variants in distinct populations is essential to validate their generalizability and potential clinical relevance (Bellenguez et al., 2022). Studying the ABI3^S209F in a Finnish cohort provides an opportunity to confirm its association with AD risk in a genetically and environmentally distinct population, thereby strengthening the evidence for its role in disease pathogenesis (Sims et al., 2017; Wightman et al., 2021).

In this study, we aim to confirm the association of ABI3^S209F with AD in the Finnish FinnGen cohort, and to investigate how this variant, alone or in combination with APOEε4, modulates cortical Aβ burden, neuroinflammation, and brain morphology in cognitively healthy older adults using positron emission tomography (PET) and magnetic resonance imaging (MRI).

Study population

The FinnGen Study is a large biobank-scale project that combines genome data and longitudinal register-based healthcare data of > 500,000 Finns (Kurki et al., 2023). In this study we used data from FinnGen release R12, which included 8,490 heterozygous carriers of ABI3^S209F and 511,670 non-carriers. For survival analysis, each ABI3^S209F carrier was assigned with up to five non-carrier controls matched for sex and the year of birth.

For the imaging part of this study, we recruited 58 subjects of ≥ 50 years of age, based on power calculations on previous studies. The participants of this cross-sectional study were recruited in collaboration with the local Auria biobank and from the Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) (Kivipelto et al., 2013) cohort. Genotype data was available for a subset of the biobank cohort through FinnGen study, allowing the biobank to directly contact persons with the ABI3^S209F and APOE ε4/ε3 or APOE ε3/ε3 genotype who had previously signed a biobank consent and an additional informed consent allowing the biobank to contact them if they are suitable for participating in a research study. Main exclusion criteria were dementia or cognitive impairment, any degenerative neurological disease, chronic inflammatory condition, and contraindication for MRI or PET imaging. Our study involved three distinct groups for comparative analysis (n = 19 + 19 + 20): 1) individuals possessing both the ABI3^S209F variant and the APOE ε4/ε3 genotype (ABI3^S209F/ε4), 2) those with the ABI3^S209F variant and the APOE ε3/ε3 genotype (ABI3^S209F/ε3), and 3) a non-carrier group with the major ABI3^S209F allele and APOE ε3/ε3 genotype (NC). The study was approved by the Ethical Committee of the Hospital District of Southwest Finland. All participants signed written informed consent.

Genotyping

ABI3^S209F is encoded by a single nucleotide substitution NM_016428.3:c.626T > C (rs616338). Although the T allele (encoding phenylalanine) is the reference allele, it is the rare allele in this locus. Thus, we refer to the common C allele (encoding serine) as the major allele and the rare T allele as the minor (effect) allele, consistent with previous reports as well as allele frequency and evolutionary conservation of the serine residue across multiple species (Sims et al., 2017).

The FinnGen cohort has been genotyped with Illumina (Illumina Inc., San Diego, USA) and Affymetrix (Thermo Fisher Scientific, Santa Clara, CA, USA) chip arrays as part of the FinnGen Study. Chip genotype data were imputed using the Finnish population-specific imputation reference panel Sequencing Initiative Suomi project (SISu v4.2, Institute for Molecular Medicine Finland, University of Helsinki, Finland, http://sisuproject.fi). Majority of the FinnGen cohort (92%) had been directly genotyped for rs616338, but for a broader coverage, imputed genotypes were used in all analyses. Imputation INFO score for rs616338 was 0.995. FINGER cohort (Kivipelto et al., 2013) was genotyped with Illumina Infinium Global Screening Array and imputed with TOPMed reference panel as described before (Céline Bellenguez et al., 2022). To confirm the ABI3 genotypes in a subset of 42 individuals enrolled to the imaging study, venous blood was collected and genomic DNA was extracted from whole blood using QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany). ABI3 rs616338 was genotyped with TaqMan assay (C_2270073_20, Applied Biosystems). All genotypes were consistent with those observed with chip genotyping or imputation.

Definition of disease endpoints

To define the disease endpoints, we utilized FinnGen core endpoints, which are based on digital health record data from Finnish health registries. Diagnoses are based on International Classification of Diseases (ICD) codes and have been harmonized over ICD-8, ICD-9 and ICD-10 codes.

AD was defined as a diagnosis in the hospital discharge or cause of death registries with the ICD codes G30 (ICD-10) or 3310 (ICD-9). AD onset age for the survival analysis was defined as the age of the first diagnosis. The remaining individuals were considered as controls. For the survival analysis, end of follow-up for controls was defined as death, moving abroad, or the latest update of the registry data, whichever came first.

Neurodegenerative disorder was defined as at least three entries with the following criteria: diagnosis in the hospital discharge or cause of death registries with the ICD codes F00*, G30 (ICD-10), 3310 (ICD-9), or 29010 (ICD-8); Finnish Social Insurance Institution (Kela) reimbursement for F00* or G30 (ICD-10) or reimbursement code 307; or prescription medicine purchases with ATC class N06D. Remaining individuals, excluding cases with AD, were considered as controls.

Dementia was defined as at least three entries with the following criteria: a diagnosis in the hospital discharge or cause of death registries with the ICD codes F00-F09 (ICD-10), 290, 3310, or 4378A (ICD-9) or 290 (ICD-8), Finnish Social Insurance Institution (Kela) reimbursement code 307, or prescription medicine purchases with ATC class N06D. Dementia cases included individuals with vascular dementia, dementia in other diseases classified elsewhere, or unspecified dementia. Remaining individuals, excluding cases with organic mental disorders, were considered as controls.

Brain imaging

All subjects underwent a structural brain MRI including T1-weighted sequences. Structural brain images were acquired with the Philips Ingenuity 3.0 T TF PET/MRI (Philips Healthcare, Amsterdam, the Netherlands). MRI was used to acquire volumetric variables for hippocampus, parahippocampus, entorhinal cortex and amygdala.

PET scans were acquired using an ECAT high-resolution research tomograph (HRRT, Siemens Medical solutions, Knoxville, TN). To estimate brain Aβ accumulation, [¹¹C]PiB scans (n = 56) were acquired 40 to 90 min post injection (mean injected dose 493 (standard deviation (SD) 42) MBq). To estimate microglial activity, we used TSPO imaging with [¹¹C]PK11195. Dynamic [¹¹C]PK11195 scans (n = 56) were acquired for 60 min post injection (mean injected dose 476 (SD 41) MBq). All images were reconstructed with 3D ordinary Poisson ordered subset expectation maximization algorithm (OP-OSEM3D), and list mode data was histogrammed into 8 (6 × 5 + 2 × 10 min, [¹¹C]PiB) and 17 (2 × 15; 3 × 30; 3 × 60; 7 × 300; 2 × 600 s, [¹¹C]PK11195) time frames.

Cognitive testing

A thorough neurocognitive test battery was administered to the participants by trained psychology students. The test battery included parts from the Finnish version of the Wechsler Memory Scale (WMS-R) and the Wechsler Adult Intelligence Scale (WAIS-R), Boston Naming Test (BNT), Trail Making Tests A and B (TMT-A and TMT-B), S-fluency, categorical fluency, and Stroop (Battery, 1944; Kaplan, Goodglass, & Weintraub, 1983; Stroop, 1935; Wechsler, 1981, 1987). Domain-specific neurocognitive test z-scores, based on an a priori hypothesis (Lezak, Howieson, Bigler, & Tranel, 2012), were calculated for executive functions, processing speed, language, and episodic memory, with higher scores indicating better performance. The executive function domain included the Trail Making Test A and B (TMT-B minus TMT-A), Stroop test (inhibition minus naming), digit span backward, and S-fluency. The processing speed domain consisted of TMT-A and digit symbol tests. The episodic memory domain included the WMS-R delayed logic memory and delayed verbal recall. The language domain included categorical fluency, the Boston naming test, and WAIS-R similarities.

Brain image analysis

Both PET and MRI images were analysed for region of interest (ROI) and voxel-wise differences between the study groups. PET and MRI image preprocessing and analysis were performed using an automated pipeline at Turku PET Centre (Karjalainen et al., 2020), which executed the PET data frame by frame realignment, PET-MRI co-registration, FreeSurfer (Freesurfer v6, https://surfer.nmr.mgh.harvard.edu/) ROI parcellation and PET data kinetic modelling. Regional and voxel level [¹¹C]PiB binding was quantified as standardized uptake value ratios (SUVR) calculated for 60 to 90 min post injection using the cerebellar cortex as the reference region. A composite neocortical [¹¹C]PiB score was calculated as the volume weighted average of the [¹¹C]PiB region-to-cerebellar cortex SUVRs for the lateral frontal, lateral temporal, and parietal cortices as well as the posterior cingulate, anterior cingulate, and precuneus. This composite [¹¹C]PiB score was used to estimate brain Aβ load. [¹¹C]PK11195-binding was quantified from the same composite region, as distribution volume ratios (DVR) within 20–60 min post injection using a reference tissue input Logan’s method with pseudo-reference region extracted using supervised clustering algorithm (Turkheimer et al., 2007; Yaqub et al., 2012). Voxel-level kinetic modeling for [¹¹C]PK11195 was carried out using basis function implementation of simplified reference tissue model with respect to the aforementioned clustered pseudo-reference region and with 300 basis functions calculated within the Θ3 parameter limits 0.06 ≤ Θ3 ≤ 0.6 (Gunn, Lammertsma, Hume, & Cunningham, 1997). Partial volume effect (PVE)-corrected data was used for all [¹¹C]PK11195 analysis to minimize the effect of PK binding in sinuses to cortical regions. PVE correction was carried out using PETPVE12 toolbox (Gonzalez-Escamilla, Lange, Teipel, Buchert, & Grothe, 2017) in both ROI (geometric transfer matrix method) and voxel-level (Muller-Gartner method) data.

The brain MRI images were also analysed for volumetric differences extracted from the FreeSurfer results. Here we focused on the hippocampus, parahippocampus, and the entorhinal cortex due to their known association with neurodegeneration related to AD (Chandra, Dervenoulas, & Politis, 2019). All voxel-wise analyses were conducted using statistical parametric mapping (SPM12 v12; Wellcome Trust Centre for Neuroimaging, London, UK) running on MATLAB R2021b (Math-Works, Natick, MA, USA).

Statistical analysis

Association of rs616338-T allele with AD, neurodegenerative disorder, and dementia were assessed in FinnGen data with chi-square test using R 4.3.2. The data are presented as OR with 95% confidence intervals (CI). Survival analysis comparing AD-free survival of ABI3^S209F carriers and age- and sex-matched controls was performed in R 4.3.2 where Kaplan-Meier curves were generated using package survminer v0.4.9 and cox proportional hazards models were fitted with survival package v3.2-7 (T Therneau, 2024; TM Therneau & Grambsch, 2000). The proportional hazards assumption was assessed using the Schoenfeld residuals test, which indicated a violation for the APOE genotype. To account for this, the Cox model was stratified by APOE genotype. The data is presented as hazard ratio (HR) with 95% CI. Results with p < 0.05 were considered statistically significant.

The sample size for the imaging part of this study was based on a power analysis calculated from previous research results obtained with the same radiotracers (Aalto et al., 2009; Z. Fan et al., 2017). The assumption was that the study would have 90% power (1–β = 0.9, α = 0.05) to detect, depending on the tracer, a 10–20% difference in regional binding ratios between the ABI3^S209F/ε4 and NC groups. The final sample size also aimed to account for potential dropouts. The ABI3^S209F/ε3 group was assumed to show an effect, with results expected to fall between those of the primary comparison groups.

Statistical analyses of the imaging data were performed using JMP Pro 16.0.0 (SAS Institute Inc., Cary, North Carolina, USA). All data following a normal distribution are presented as mean (SD), otherwise as median (interquartile range, IQR). The normality of the data was evaluated visually from the distribution and with the Shapiro-Wilk test. Differences in continuous variables between the three groups were tested using linear regression models, adjusting for age and sex. The MRI variables were also adjusted for total intracranial volume. If a significant effect was found, all pairs were compared using the post hoc Tukey’s honest significance test for multiple comparisons.

Volumetric differences in T1 MRI, [¹¹C]PIB, and [¹¹C]PK11195-binding at the voxel level were assessed using ANCOVA, using age and sex as covariates for [¹¹C]PIB and [¹¹C]PK11195, but also total intracranial volume for MRI. This was followed by post-hoc pairwise comparisons in SPM12. Voxel-wise differences between the groups were tested using linear regression in SPM12 to evaluate if differences were present also outside the a priori chosen brain regions. Uncorrected p < 0.001 combined with a cluster-level false discovery rate (FDR) correction for multiple comparisons was considered statistically significant in the voxel-based analyses. When significant FDR corrected clusters were found we applied family wise error (FWE) correction with p < 0.05 to see if the results survived the tighter threshold.

To combine the scores of different neurocognitive tests into domain-specific scores, z-scores for the tests were calculated by standardizing the raw test scores to the study population’s mean and standard deviation. To achieve a normal distribution, one outlier per raw test were excluded before the z-transformation (1 outlier in Boston naming test; and 1 outlier in Stroop test). All outliers performed worse on the cognitive tests than − 2 SD of the present study population. A skewness of − 1 to 1 was accepted for the raw test score distribution. Domain-specific z-scores were determined by averaging all the z-scores within each domain. For tests with a reverse scale (Stroop, TMT-A, and TMT-B), reciprocal numbers were used to ensure higher scores indicated better performance. Participants with missing test results were excluded.

ABI3^S209F associates with increased risk and earlier onset age of AD in FinnGen cohort

ABI3^S209F (rs616338-T) was significantly associated with increased risk of AD (OR = 1.22, 95% CI: 1.09–1.38, p = 0.0012) and neurodegenerative disorder (OR = 1.21, 95% CI: 1.10–1.34, p = 0.00023) in FinnGen (Fig. 1A). Association with dementia was not statistically significant (OR = 1.10, 95% CI: 1.00-1.20, p = 0.06).

To investigate the impact of ABI3^S209F and APOE ε4 on the age of onset of AD, we performed survival analysis for age at AD diagnosis. Kaplan-Meier survival curves indicated an earlier onset of AD among the ABI3^S209F carriers when compared to the non-carriers with the same APOE genotype (Fig. 1B). Stratified Cox proportional hazards models were used to account for the violation of the proportional hazard assumption by APOE genotype. Within the APOE ε3 stratum, ABI3^S209F carriers exhibited a significantly increased hazard of AD onset (HR = 1.31, 95% CI: 1.09–1.59, p = 0.0046). Similarly, in the APOE ε4 stratum, ABI3^S209F carriers showed an elevated risk (HR = 1.32, 95% CI: 1.09–1.60, p = 0.0045). These findings suggest that the ABI3^S209F variant is associated with earlier AD onset, independent of APOE genotype.

Demographics of the PET–MRI subjects

Baseline demographic, clinical, and imaging characteristics of the PET–MRI subsample are presented in Table 1. Across the three genotype-defined study groups (ABI3^S209F/ε4, ABI3^S209F/ε3, and NC), no statistically significant differences were observed in sex distribution, age, height, or hippocampal volume (P > 0.07 for all). Cognitive performance, as assessed by MMSE, was similar across groups. Importantly, a significant group difference was observed in cortical Aβ burden, estimated as composite [¹¹C]PiB SUVR, with the ABI3^S209F/ε3 group showing a lower median binding score than the ABI3^S209F/ε4 group (p = 0.01). No differences were observed in microglial activation ([¹¹C]PK11195 DVR).

Table 1

Subject demographics and descriptive data
		ABI3^S209F/ε4	ABI3^S209F/ε3	NC
N		19	19	20
Sex	female	11	8	11	0.58
	male	8	11	9
Age	Mean (sd)	69.5 (6.2)	71.4 (6.9)	71.4 (5.1)	0.55
Height	Mean (sd)	168.9 (9.4)	172.2 (9.6)	168.3 (7.9)	0.39
Weight	Mean (sd)	81.8 (12.7)	83.4 (18)	73.1 (13.2)	0.073
BMI	Mean (sd)	28.8 (4.5)	27.9 (4.2)	25.7 (3.7)	0.066
[¹¹C]PiB score	Mean (sd))	2.0 (0.7)	1.67 (0.5)*	1.75 (0.6)	0.01
[¹¹C]PK DVR	Mean (sd)	1.3 (0.1)	1.3 (0.1)	1.3 (0.1)	0.83
Hippocampal volume	Mean (sd)	7.7 (0.8)	7.7 (1.0)	7.2 (0.6)	0.26
MMSE	Median (IQR)	28.5 (28–29)	29 (28–29)	29 (28–30)	0.83
Data are presented as mean (standard deviation) or median (interquartile range) depending on the distribution. Differences between groups were tested with ANCOVA with Tukey’s honest significance test, Kruskal–Wallis test with Dunn’s method for multiple or linear models. χ2 test was used for testing categorical variables. P value presents overall difference between groups. Significant differences in pairwise comparisons to ABI+/e4e3 (p < 0.05 = *) are presented.

APOEε4, but not ABI3^S209F, increases brain Aβ load

In the ROI analysis, the ABI3^S209F/ε4 had higher composite cortical [¹¹C]PiB score when compared to the ABI3^S209F/ε3 (2.0 (0.7) vs. 1.67 (0.5); mean (sd), p = 0.017 (HST)) (Fig. 2a). The voxel-level analysis showed higher [¹¹C]PiB binding across the cortex in the ABI3^S209F/ε4 group than in the two other groups (Fig. 3).

No global, but subtle regional neuroinflammatory differences in ABI3^S209F/ε4 carriers.

The [¹¹C]PK11195 analysis revealed no significant differences between the groups in the composite cortical ROI (Fig. 2b). However, voxel-level analysis detected minor differences between the ABI3^S209F/ε4 and both the ABI3^S209F/ε3 and NC groups, but not between the ABI3^S209F/ε3 and NC groups (Fig. 4). The ABI3^S209F/ε4 showed higher [¹¹C]PK11195 binding in the precuneus and parieto-occipital regions compared to the ABI3^S209F/ε3 group. Additionally, higher [¹¹C]PK11195 binding was observed in the parieto-occipital and calcarine regions of the ABI3^S209F/ε4 compared to the NC group.

While no significant ROI-level correlations were found between the PET tracers, some voxel-wise associations were identified mainly in subcortical regions in the white matter. In addition, a weak correlation (r = 0.27, p = 0.043) was observed between [¹¹C]PK11195 binding and cortical volume. Full correlation results are provided in the Supplementary Materials (Table S1, Figure S1).

ABI3^S209F does not significantly affect brain region volumes

In the prechosen ROIs we found no differences between the groups in the volumes of cerebral cortex, hippocampus, parahippocampus, entorhinal cortex or amygdala (Supplementary material, Table S2). Subsequent VBM analysis found that the ABI3^S209F/ε3 had lower volumes than the other groups mainly in the superior temporal gyrus, when correcting for FDR, but these findings did not survive the FEW correction (Supplementary material, Figure S2).

ABI3^S209F has no effect on cognition

Global cognition was assessed with the MMSE, and domain-specific performance was summarized as z-standardized composite. One-way ANOVA followed by Tukey’s honest significant difference test revealed no overall between-group differences for MMSE (Table 1) or any cognitive domain (Table 2, Fig. 5).

Table 2

Cognitive test results
		ABI3^S209F/ε4	ABI3^S209F/ε3	NC	Group difference
Executive functions	Mean (sd)	0.041 (0.41)	-0.046 (0.61)	0.035 (0.47)	0.47
Processing speed	Mean (sd)	0.024 (0.79)	0.30 (0.86)	0.21 (0.87)	0.85
Episodic memory	Mean (sd)	-0.34 (0.73)	-0.11 (0.65)	-0.10 (0.52)	0.63
Language functions	Mean (sd)	-0.16 (0.58)	0.12 (0.66)	0.14 (0.53)	0.23
Data are presented as mean (standard deviation). Differences between groups were tested with one-way ANOVA with Tukey’s honest significance test. P value presents overall difference between groups. Significant differences in pairwise comparisons to ABI+/e4e3 (p < 0.05 = *) are presented

This study investigated the ABI3^S209F genetic variant as a risk factor for AD using data from the large Finnish FinnGen cohort. In addition, we experimentally assessed its impact, alone or combined with the APOEε4, on cortical Aβ deposition, neuroinflammation, brain morphology, and cognitive performance in a separate subsample of cognitively healthy older adults.

Genetic analysis confirmed ABI3^S209F as a significant AD risk variant (OR = 1.22), though with a somewhat smaller effect size compared to previous findings (Conway et al., 2018; Dalmasso et al., 2019; Olive et al., 2020; Satoh et al., 2017; Sims et al., 2017; Wightman et al., 2021). Additionally, survival analyses indicated that ABI3^S209F carriers developed AD earlier than non-carriers. In the experimental analyses, ABI3^S209F alone did not influence cortical Aβ accumulation, neuroinflammation or regional brain volumes. However, subtle regional effects emerged in voxel-wise analyses, particularly for ABI3^S209F combined with APOEε4, suggesting region-specific Aβ deposition and modest neuroinflammation not detected by broader ROI analyses. Collectively, these findings replicate and extend prior results, highlighting the complexity and context-dependent nature of ABI3^S209F’s contribution to AD pathology.

In a genetic association analysis in FinnGen, the ABI3^S209F variant significantly associated with AD and more broadly defined neurodegenerative disorder, with comparable ORs across the endpoints. Interestingly, no significant association was observed with the broadest endpoint definition, dementia. This likely reflects differences in definitions of the broader endpoints: while the neurodegenerative disorder endpoint may more specifically capture AD and related pathologies, the dementia endpoint encompasses a heterogeneous group of conditions, including vascular and unspecified dementias. The lack of association with dementia may therefore result from dilution of AD-specific genetic signals within this phenotypically diverse group.

Beyond AD risk, ABI3^S209F also appears to influence the timing of disease onset. Survival analysis revealed that carriers of the variant developed AD at an earlier age compared to non-carriers, with consistent effect observed across APOE ε3/ε3 and ε3/ε4 backgrounds. These findings suggest that ABI3^S209F may act as a disease modifier, accelerating the onset of AD independently of APOE genotype. Taken together, these results reinforce the specificity of the association of ABI3^S209F variant with AD and highlight its potential role in modulating disease progression.

In the imaging part of the study, we examined the effect of ABI3^S209F, in the presence or absence of the APOEε4, on brain Aβ deposition, neuroinflammation, brain structures, and cognitive functions in cognitively healthy older adults. Although ABI3^S209F has previously been associated with increased AD risk in several studies, including replication in our current study, we did not observe increased cortical Aβ deposition in ABI3^S209F/ε3 carriers when compared to NC, either in ROI-based or voxel-wise analyses. This suggests that ABI3^S209F alone, without the APOE ε4 allele, may have limited functional impact on Aβ accumulation in cognitively normal individuals.

Interestingly, the highest cortical Aβ levels in ROI analysis were observed in participants carrying both ABI3^S209F and APOEε4. However, these were only significantly different compared to the ABI3^S209F/ε3 group, not to NC. Voxel-wise analysis, in contrast, revealed region-specific increases in [¹¹C]PiB signal in the ABI3^S209F/ε4 group compared to both other groups, particularly in frontal, temporal, and posterior cortices. Previous work has shown that one APOEε4 alone does not consistently induce such voxel-level differences (Snellman et al., 2023; Reiman et al., 2009), further supporting a gene–gene interaction hypothesis. The ROI findings align with previous reports showing no significant association between ABI3^S209F and CSF Aβ42 levels or Aβ-PET signal in cognitively healthy individuals (Olive et al., 2020). The voxel-level differences suggest that the combination of ABI3^S209F and APOEε4 may lead to regionally specific Aβ accumulation not captured by conventional ROI analyses, supporting a synergistic gene–gene interaction. Animal studies have yielded mixed results regarding ABI3’s role in Aβ pathology. While ABI3 knockout has been associated with increased Aβ deposition in some studies (Karahan et al., 2021), others have reported reduced deposition, particularly at early time points (Ibanez et al., 2022). In the latter case, the reduction was transient, and Aβ levels eventually caught up with NC as the animals aged.

To our knowledge, this is the first study to examine TSPO PET imaging in cognitively normal individuals carrying the ABI3^S209F variant. Previous studies with various PET-ligands have reported increased TSPO binding between AD and NC (Cagnin et al., 2001; Edison et al., 2008; Zhen Fan, Okello, Brooks, & Edison, 2015), and also in Aβ-positive MCI (Okello et al., 2009; Parbo et al., 2017) and Aβ-positive NC (Z. Fan et al., 2017; Zou et al., 2020). However, when comparing a composite ROI covering the entire cortical grey matter, we found no significant differences in [¹¹C]PK11195 binding between the genotype groups, although Aβ accumulation was clearly highest in the ABI3^S209F/ε4 group in the same region. Voxel-wise analysis revealed a subtle increase in [¹¹C]PK11195 binding in the parieto-occipital cortex of the ABI3^S209F/ε4 group compared to the other groups. Although this region also showed elevated Aβ signal, the highest Aβ accumulation was observed in the frontal cortex, where no corresponding increase in [¹¹C]PK11195 binding was detected. Our findings suggest that while the combination of ABI3^S209F and APOEε4 is associated with increased cortical Aβ accumulation, this does not correspond to a parallel increase in microglial activation as measured by [¹¹C]PK11195. A similar dissociation has previously been reported among APOEε4 carriers with varying allele loads (Snellman et al., 2023). This mismatch may reflect the limited sensitivity of first-generation TSPO tracers such as [¹¹C]PK11195, which are known to suffer from poor signal-to-noise ratio and high nonspecific binding (Vivash & O'Brien, 2016; Yokokura et al., 2017) or a temporal lag between Aβ deposition and immune response, or the functional heterogeneity of microglial activation, which is not fully captured by TSPO PET imaging (Z. Fan et al., 2017). Overall, while the voxel- and ROI-level data diverge in some respects, both support a nuanced and genotype-dependent role for ABI3^S209F in AD pathology.

The strength of this study is the combination of population-scale genetic data with in vivo multimodal imaging, enabling a rare genotype-specific characterization of early AD-related pathology in cognitively normal individuals. The use of the large and well-characterized FinnGen cohort allowed us to validate the ABI3^S209F AD association in a genetically distinct Northern European population, adding valuable replication evidence to earlier findings. A major practical strength of this study was the ability to efficiently recruit participants with specific genotypes through biobank collaboration. Without access to genotype-based preselection, recruiting a sufficient number of ABI3^S209F carriers would have required genotyping and screening several hundred individuals. The biobank-based recruitment thus greatly increased feasibility and resource efficiency. Additionally, the integration of PET imaging for both Aβ and microglial activity, alongside MRI-derived structural metrics and neuropsychological testing, enabled a multidimensional assessment of functional consequences at an early disease stage.

However, some limitations should be noted. The use of the first-generation TSPO tracer [¹¹C]PK11195 may have constrained sensitivity to microglial activation due to its limited signal-to-noise ratio and high nonspecific binding. Also, the sample size in our imaging part was modest, although it was based on power calculations on previous studies there is a chance of type II error. Furthermore, although voxel-wise analysis increases spatial precision, it also introduces multiple comparison challenges and increases the risk of type I error, despite statistical correction. Finally, while the study design allowed for the evaluation of ABI3^S209F effects in isolation and in combination with APOEε4, the cross-sectional nature of the imaging component precludes firm conclusions about temporal dynamics or causality.

Despite these limitations, the present findings highlight the importance of considering gene–gene interactions and regional brain vulnerability in understanding AD risk. Further studies using larger imaging cohorts, more sensitive second-generation tracers, and longitudinal follow-up will be essential to clarify the mechanisms by which ABI3 and APOEε4 interact to influence early neuropathological changes.

This study confirms the association of the ABI3^S209F variant with increased risk for AD. The variant was also associated with an earlier age of AD onset. ABI3^S209F alone does not significantly affect cortical Aβ deposition or neuroinflammation. However, the combination of ABI3^S209F and APOEε4 may contribute to region-specific Aβ accumulation and increased microglial activation. Our findings emphasize the complex and context-dependent role of ABI3^S209F in AD pathophysiology, possibly acting as a modifier rather than a primary driver of pathology. The lack of consistent differences in cognitive performance and brain volumes supports the notion that ABI3^S209F-related changes may precede symptomatic disease or require additional genetic or environmental interactions to manifest. Further longitudinal and mechanistic studies are needed to elucidate the functional consequences of this variant and its potential as a therapeutic target or biomarker in preclinical stages of AD.

Alzheimer’s disease

ApoE

Apolipoprotein E

APOEε4

APOE ε4 allelle

Aβ

Beta-amyloid

ABI3

Abelson interactor family member 3

ABI3^S209F

ABI3 p.Ser209Phe variant

APP

Amyloid precursor protein

ATC

Anatomical Therapeutic Chemical

BNT

Boston Naming Test

CSF

Cerebrospinal fluid

DVR

Distribution Volume Ratio

FDR

False discovery rate

FINGER

Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability

FinnGen

Finnish genetic research project combining genomic and health registry data

FWE

Family-wise error

GWAS

Genome-wide association study

HRRT

High-resolution research tomograph

ICD

International Classification of Diseases

IQR

Interquartile range

MBq

Megabecquerel

MCI

Mild cognitive impairment

MRI

Magnetic resonance imaging

PET

Positron emission tomography

PVE

Partial volume effect

ROI

Region of interest

Standard deviation

SPM

Statistical parametric mapping

SUVR

Standardized uptake value ratio

TSPO

Translocator protein

Ethics approval and consent to participate

The brain imaging study was approved by the Ethical Committee of the Hospital District of Southwest Finland. Biobank participant recontacting was approved by the Scientific Steering Committee of Auria Biobank. All participants signed a written informed consent.

The study subjects in FinnGen provided informed consent for biobank research, based on the Finnish Biobank Act. Alternatively, separate research cohorts, collected prior the Finnish Biobank Act came into effect (in September 2013) and start of FinnGen (August 2017), were collected based on study-specific consents and later transferred to the Finnish biobanks after approval by Fimea (Finnish Medicines Agency), the National Supervisory Authority for Welfare and Health. Recruitment protocols followed the biobank protocols approved by Fimea. The Coordinating Ethics Committee of the Hospital District of Helsinki and Uusimaa (HUS) statement number for the FinnGen study is Nr HUS/990/2017.

The FinnGen study is approved by Finnish Institute for Health and Welfare (permit numbers: THL/2031/6.02.00/2017, THL/1101/5.05.00/2017, THL/341/6.02.00/2018, THL/2222/6.02.00/2018, THL/283/6.02.00/2019, THL/1721/5.05.00/2019 and THL/1524/5.05.00/2020), Digital and population data service agency (permit numbers: VRK43431/2017-3, VRK/6909/2018-3, VRK/4415/2019-3), the Social Insurance Institution (permit numbers: KELA 58/522/2017, KELA 131/522/2018, KELA 70/522/2019, KELA 98/522/2019, KELA 134/522/2019, KELA 138/522/2019, KELA 2/522/2020, KELA 16/522/2020), Findata permit numbers THL/2364/14.02/2020, THL/4055/14.06.00/2020, THL/3433/14.06.00/2020, THL/4432/14.06/2020, THL/5189/14.06/2020, THL/5894/14.06.00/2020, THL/6619/14.06.00/2020, THL/209/14.06.00/2021, THL/688/14.06.00/2021, THL/1284/14.06.00/2021, THL/1965/14.06.00/2021, THL/5546/14.02.00/2020, THL/2658/14.06.00/2021, THL/4235/14.06.00/2021, Statistics Finland (permit numbers: TK-53-1041-17 and TK/143/07.03.00/2020 (earlier TK-53-90-20) TK/1735/07.03.00/2021, TK/3112/07.03.00/2021) and Finnish Registry for Kidney Diseases permission/extract from the meeting minutes on 4th July 2019.

The Biobank Access Decisions for FinnGen samples and data utilized in FinnGen Data Freeze 12 include: THL Biobank BB2017_55, BB2017_111, BB2018_19, BB_2018_34, BB_2018_67, BB2018_71, BB2019_7, BB2019_8, BB2019_26, BB2020_1, BB2021_65, Finnish Red Cross Blood Service Biobank 7.12.2017, Helsinki Biobank HUS/359/2017, HUS/248/2020, HUS/430/2021 §28, §29, HUS/150/2022 §12, §13, §14, §15, §16, §17, §18, §23, §58, §59, HUS/128/2023 §18, Auria Biobank AB17-5154 and amendment #1 (August 17 2020) and amendments BB_2021-0140, BB_2021-0156 (August 26 2021, Feb 2 2022), BB_2021-0169, BB_2021-0179, BB_2021-0161, AB20-5926 and amendment #1 (April 23 2020) and it´s modifications (Sep 22 2021), BB_2022-0262, BB_2022-0256, Biobank Borealis of Northern Finland_2017_1013, 2021_5010, 2021_5010 Amendment, 2021_5018, 2021_5018 Amendment, 2021_5015, 2021_5015 Amendment, 2021_5015 Amendment_2, 2021_5023, 2021_5023 Amendment, 2021_5023 Amendment_2, 2021_5017, 2021_5017 Amendment, 2022_6001, 2022_6001 Amendment, 2022_6006 Amendment, 2022_6006 Amendment, 2022_6006 Amendment_2, BB22-0067, 2022_0262, 2022_0262 Amendment, Biobank of Eastern Finland 1186/2018 and amendment 22§/2020, 53§/2021, 13§/2022, 14§/2022, 15§/2022, 27§/2022, 28§/2022, 29§/2022, 33§/2022, 35§/2022, 36§/2022, 37§/2022, 39§/2022, 7§/2023, 32§/2023, 33§/2023, 34§/2023, 35§/2023, 36§/2023, 37§/2023, 38§/2023, 39§/2023, 40§/2023, 41§/2023, Finnish Clinical Biobank Tampere MH0004 and amendments (21.02.2020 & 06.10.2020), BB2021-0140 8§/2021, 9§/2021, §9/2022, §10/2022, §12/2022, 13§/2022, §20/2022, §21/2022, §22/2022, §23/2022, 28§/2022, 29§/2022, 30§/2022, 31§/2022, 32§/2022, 38§/2022, 40§/2022, 42§/2022, 1§/2023, Central Finland Biobank 1-2017, BB_2021-0161, BB_2021-0169, BB_2021-0179, BB_2021-0170, BB_2022-0256, BB_2022-0262, BB22-0067, Decision allowing to continue data processing until 31st Aug 2024 for projects: BB_2021-0179, BB22-0067,BB_2022-0262, BB_2021-0170, BB_2021-0164, BB_2021-0161, and BB_2021-0169, and Terveystalo Biobank STB 2018001 and amendment 25th Aug 2020, Finnish Hematological Registry and Clinical Biobank decision 18th June 2021, Arctic biobank P0844: ARC_2021_1001.

Data availability

The brain imaging datasets used and analysed during the current study are available from the corresponding author on reasonable request. Summary statistics from FinnGen data release 12 are publicly available and can be accessed at https://www.finngen.fi/en/access_results and https://r12.finngen.fi. Access for individual level genotype data can be applied for via the Fingenious portal (https://site.fingenious.fi/en/) hosted by the Finnish Biobank Cooperative FinBB (https://finbb.fi/en/). Access to Finnish Health register data can be applied from Findata (https://findata.fi/en/data/).

Competing interests

Funding

MK was supported by Yrjö Jahnsson Foundation. HM and MH were supported by the Research Council of Finland (grants #355604, #338182). MH was supported by the Sigrid Jusélius Foundation, Jane and Aatos Erkko Foundation, the Strategic Neuroscience Funding of the University of Eastern Finland, Faculty of Health Sciences of University of Eastern Finland, Alzheimer's Association (ADSF-24-1284326-C and AARG-22-926152).

FinnGen is funded by two grants from Business Finland (HUS 4685/31/2016 and UH 4386/31/2016) and the following industry partners: AbbVie Inc., AstraZeneca UK Ltd, Biogen MA Inc., Bristol Myers Squibb (and Celgene Corporation & Celgene International II Sàrl), Genentech Inc., Merck Sharp & Dohme LCC, Pfizer Inc., GlaxoSmithKline Intellectual Property Development Ltd., Sanofi US Services Inc., Maze Therapeutics Inc., Janssen Biotech Inc, Novartis Pharma AG, and Boehringer Ingelheim International GmbH.

Authors' contributions

MK, JOR and HM designed the study concept. MK drafted the manuscript. MK contributed to data collection. MK and HM analysed the data. MK, JOR, MT, AS and HM contributed to study design and interpretation of the data. JOR, AS, and MH supervised the study. All authors read and critically revised the manuscript for its content and approved the final submitted version.

Acknowledgements

Authors would like to thank Dr. Christian Haass (DZNE, Germany) for his valuable comments on the manuscript. We thank the participants and researchers of FinnGen. We thank Merja Perälä from Auria Biobank for her valuable work in recruiting the subjects.

Aalto, S., Scheinin, N. M., Kemppainen, N. M., Någren, K., Kailajärvi, M., Leinonen, M., . . . Rinne, J. O. (2009). Reproducibility of automated simplified voxel-based analysis of PET amyloid ligand [11C]PIB uptake using 30-min scanning data. European Journal of Nuclear Medicine and Molecular Imaging, 36(10), 1651-1660. doi:10.1007/s00259-009-1174-1
Battery, A. I. T. (1944). Manual of Directions and Scoring. In. Washington DC: War Department, Adjutant General’s Ofﬁce.
Bellenguez, C., Küçükali, F., Jansen, I. E., Kleineidam, L., Moreno-Grau, S., Amin, N., . . . Lambert, J.-C. (2022). New insights into the genetic etiology of Alzheimer’s disease and related dementias. Nature Genetics, 54(4), 412-436. doi:10.1038/s41588-022-01024-z
Bellenguez, C., Küçükali, F., Jansen, I. E., Kleineidam, L., Moreno-Grau, S., Amin, N., . . . Lambert, J. C. (2022). New insights into the genetic etiology of Alzheimer's disease and related dementias. Nat Genet, 54(4), 412-436. doi:10.1038/s41588-022-01024-z
Cagnin, A., Brooks, D. J., Kennedy, A. M., Gunn, R. N., Myers, R., Turkheimer, F. E., . . . Banati, R. B. (2001). In-vivo measurement of activated microglia in dementia. The Lancet (British edition), 358(9280), 461-467. doi:10.1016/S0140-6736(01)05625-2
Castellano, J. M., Kim, J., Stewart, F. R., Jiang, H., Demattos, R. B., Patterson, B. W., . . . Holtzman, D. M. (2011). Human apoE Isoforms Differentially Regulate Brain Amyloid-β Peptide Clearance. Science Translational Medicine, 3(89), 89ra57-89ra57. doi:10.1126/scitranslmed.3002156
Castillo, E., Leon, J., Mazzei, G., Abolhassani, N., Haruyama, N., Saito, T., . . . Nakabeppu, Y. (2017). Comparative profiling of cortical gene expression in Alzheimer’s disease patients and mouse models demonstrates a link between amyloidosis and neuroinflammation. Scientific Reports, 7(1). doi:10.1038/s41598-017-17999-3
Chandra, A., Dervenoulas, G., & Politis, M. (2019). Magnetic resonance imaging in Alzheimer’s disease and mild cognitive impairment. Journal of Neurology, 266(6), 1293-1302. doi:10.1007/s00415-018-9016-3
Conway, O. J., Carrasquillo, M. M., Wang, X., Bredenberg, J. M., Reddy, J. S., Strickland, S. L., . . . Ertekin-Taner, N. (2018). ABI3 and PLCG2 missense variants as risk factors for neurodegenerative diseases in Caucasians and African Americans. Molecular Neurodegeneration, 13(1). doi:10.1186/s13024-018-0289-x
Corder, E. H., Saunders, A. M., Strittmatter, W. J., Schmechel, D. E., Gaskell, P. C., Small, G. W., . . . Pericak-Vance, M. A. (1993). Gene Dose of Apolipoprotein E Type 4 Allele and the Risk of Alzheimer's Disease in Late Onset Families. Science (American Association for the Advancement of Science), 261(5123), 921-923. doi:10.1126/science.8346443
Dalmasso, M. C., Brusco, L. I., Olivar, N., Muchnik, C., Hanses, C., Milz, E., . . . Ramirez, A. (2019). Transethnic meta-analysis of rare coding variants in PLCG2, ABI3, and TREM2 supports their general contribution to Alzheimer’s disease. Translational Psychiatry, 9(1). doi:10.1038/s41398-019-0394-9
Davidson, A. J., Ura, S., Thomason, P. A., Kalna, G., & Insall, R. H. (2013). Abi is required for modulation and stability but not localization or activation of the SCAR/WAVE complex. Eukaryot Cell, 12(11), 1509-1516. doi:10.1128/ec.00116-13
Edison, P., Archer, H. A., Gerhard, A., Hinz, R., Pavese, N., Turkheimer, F. E., . . . Brooks, D. J. (2008). Microglia, amyloid, and cognition in Alzheimer's disease: An [11C](R)PK11195-PET and [11C]PIB-PET study. Neurobiol Dis, 32(3), 412-419. doi:10.1016/j.nbd.2008.08.001
Fan, Z., Brooks, D. J., Okello, A., & Edison, P. (2017). An early and late peak in microglial activation in Alzheimer's disease trajectory. Brain, 140(3), 792-803. doi:10.1093/brain/aww349
Fan, Z., Okello, A. A., Brooks, D. J., & Edison, P. (2015). Longitudinal influence of microglial activation and amyloid on neuronal function in Alzheimer's disease. Brain (London, England : 1878), 138(Pt 12), 3685-3698. doi:10.1093/brain/awv288
Gonzalez-Escamilla, G., Lange, C., Teipel, S., Buchert, R., & Grothe, M. J. (2017). PETPVE12: an SPM toolbox for Partial Volume Effects correction in brain PET – Application to amyloid imaging with AV45-PET. NeuroImage (Orlando, Fla.), 147, 669-677. doi:10.1016/j.neuroimage.2016.12.077
Guerreiro, R. J., Gustafson, D. R., & Hardy, J. (2012). The genetic architecture of Alzheimer's disease: beyond APP, PSENs and APOE. Neurobiol Aging, 33(3), 437-456. doi:10.1016/j.neurobiolaging.2010.03.025
Gunn, R. N., Lammertsma, A. A., Hume, S. P., & Cunningham, V. J. (1997). Parametric Imaging of Ligand-Receptor Binding in PET Using a Simplified Reference Region Model. NeuroImage (Orlando, Fla.), 6(4), 279-287. doi:10.1006/nimg.1997.0303
Ibanez, K. R., Mcfarland, K. N., Phillips, J., Allen, M., Lessard, C. B., Zobel, L., . . . Chakrabarty, P. (2022). Deletion of Abi3/Gngt2 influences age-progressive amyloid β and tau pathologies in distinctive ways. Alzheimer's Research & Therapy, 14(1). doi:10.1186/s13195-022-01044-1
Jansen, I. E., Savage, J. E., Watanabe, K., Bryois, J., Williams, D. M., Steinberg, S., . . . Posthuma, D. (2019). Genome-wide meta-analysis identifies new loci and functional pathways influencing Alzheimer's disease risk. Nat Genet, 51(3), 404-413. doi:10.1038/s41588-018-0311-9
Kaplan, E., Goodglass, H., & Weintraub, S. (1983). The Boston Naming Test. Philadelphia: Lea & Febiger.
Karahan, H., Smith, D. C., Kim, B., Dabin, L. C., Al-Amin, M. M., Wijeratne, H. R. S., . . . Kim, J. (2021). Deletion of ABI3 gene locus exacerbates neuropathological features of Alzheimer’s disease in a mouse model of Aβ amyloidosis. Science Advances, 7(45). doi:10.1126/sciadv.abe3954
Karahan, H., Smith, D. C., Kim, B., Mccord, B., Mantor, J., John, S. K., . . . Kim, J. (2023). The effect of Abi3 locus deletion on the progression of Alzheimer’s disease-related pathologies. Frontiers in Immunology, 14. doi:10.3389/fimmu.2023.1102530
Karjalainen, T., Tuisku, J., Santavirta, S., Kantonen, T., Bucci, M., Tuominen, L., . . . Nummenmaa, L. (2020). Magia: Robust Automated Image Processing and Kinetic Modeling Toolbox for PET Neuroinformatics. Front Neuroinform, 14, 3. doi:10.3389/fninf.2020.00003
Kivipelto, M., Solomon, A., Ahtiluoto, S., Ngandu, T., Lehtisalo, J., Antikainen, R., . . . Soininen, H. (2013). The Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER): Study design and progress. Alzheimer's & Dementia, 9(6), 657-665. doi:10.1016/j.jalz.2012.09.012
Krasemann, S., Madore, C., Cialic, R., Baufeld, C., Calcagno, N., El Fatimy, R., . . . Butovsky, O. (2017). The TREM2-APOE Pathway Drives the Transcriptional Phenotype of Dysfunctional Microglia in Neurodegenerative Diseases. Immunity, 47(3), 566-581.e569. doi:10.1016/j.immuni.2017.08.008
Kurki, M. I., Karjalainen, J., Palta, P., Sipilä, T. P., Kristiansson, K., Donner, K. M., . . . Palotie, A. (2023). FinnGen provides genetic insights from a well-phenotyped isolated population. Nature, 613(7944), 508-518. doi:10.1038/s41586-022-05473-8
Leng, F., & Edison, P. (2021). Neuroinflammation and microglial activation in Alzheimer disease: where do we go from here? Nature Reviews Neurology, 17(3), 157-172. doi:10.1038/s41582-020-00435-y
Lezak, M., Howieson, D., Bigler, E., & Tranel, D. (2012). Neuropsychological Assessment. In. New York: Oxford Univer sity Press
Liu, C.-C., Wang, N., Chen, Y., Inoue, Y., Shue, F., Ren, Y., . . . Bu, G. (2023). Cell-autonomous effects of APOE4 in restricting microglial response in brain homeostasis and Alzheimer’s disease. Nature Immunology, 24(11), 1854-1866. doi:10.1038/s41590-023-01640-9
Nguyen, A. T., Wang, K., Hu, G., Wang, X., Miao, Z., Azevedo, J. A., . . . Lee, E. B. (2020). APOE and TREM2 regulate amyloid-responsive microglia in Alzheimer’s disease. Acta Neuropathologica, 140(4), 477-493. doi:10.1007/s00401-020-02200-3
Okello, A., Edison, P., Archer, H. A., Turkheimer, F. E., Kennedy, J. C., Bullock, R., . . . Brooks, D. J. (2009). Microglial activation and amyloid deposition in mild cognitive impairment A PET study. doi:10.1212/01.wnl.0000338622.27876.0d
Olive, C., Ibanez, L., Farias, F. H. G., Wang, F., Budde, J. P., Norton, J. B., . . . Fernandez, M. V. (2020). Examination of the Effect of Rare Variants in TREM2, ABI3, and PLCG2 in LOAD Through Multiple Phenotypes. Journal of Alzheimer's Disease, 77(4), 1469-1482. doi:10.3233/jad-200019
Parbo, P., Ismail, R., Hansen, K. V., Amidi, A., Mårup, F. H., Gottrup, H., . . . Brooks, D. J. (2017). Brain inflammation accompanies amyloid in the majority of mild cognitive impairment cases due to Alzheimer's disease. Brain, 140(7), 2002-2011. doi:10.1093/brain/awx120
Reiman, E. M., Chen, K., Liu, X., Bandy, D., Yu, M., Lee, W., . . . Caselli, R. J. (2009). Fibrillar amyloid-β burden in cognitively normal people at 3 levels of genetic risk for Alzheimer's disease. Proceedings of the National Academy of Sciences, 106(16), 6820-6825. doi:10.1073/pnas.0900345106
Rodriguez, G. A., Tai, L. M., Ladu, M. J., & Rebeck, G. W. (2014). Human APOE4 increases microglia reactivity at Aβ plaques in a mouse model of Aβ deposition. Journal of Neuroinflammation, 11(1), 111. doi:10.1186/1742-2094-11-111
Satoh, J.-I., Kino, Y., Yanaizu, M., Tosaki, Y., Sakai, K., Ishida, T., & Saito, Y. (2017). Microglia express ABI3 in the brains of Alzheimer's disease and Nasu-Hakola disease. Intractable & Rare Diseases Research, 6(4), 262-268. doi:10.5582/irdr.2017.01073
Sekino, S., Kashiwagi, Y., Kanazawa, H., Takada, K., Baba, T., Sato, S., . . . Tani, K. (2015). The NESH/Abi-3-based WAVE2 complex is functionally distinct from the Abi-1-based WAVE2 complex. Cell Communication and Signaling, 13(1). doi:10.1186/s12964-015-0119-5
Sims, R., Van Der Lee, S. J., Naj, A. C., Bellenguez, C., Badarinarayan, N., Jakobsdottir, J., . . . Schellenberg, G. D. (2017). Rare coding variants in PLCG2, ABI3, and TREM2 implicate microglial-mediated innate immunity in Alzheimer's disease. Nature Genetics, 49(9), 1373-1384. doi:10.1038/ng.3916
Snellman, A., Ekblad, L. L., Tuisku, J., Koivumäki, M., Ashton, N. J., Lantero-Rodriguez, J., . . . Rinne, J. O. (2023). APOE ε4 gene dose effect on imaging and blood biomarkers of neuroinflammation and beta-amyloid in cognitively unimpaired elderly. Alzheimer's Research & Therapy, 15(1). doi:10.1186/s13195-023-01209-6
Stroop, J., Ridley,. (1935). Studies of interference in serial verbal reactions. In (Vol. 18, pp. 643–662): Journal of Experimental Psychology.
Therneau, T. (2024). A package for survival analysis in R. Retrieved from https://cran.r-project.org/web/packages/survival/vignettes/survival.pdf
Therneau, T., & Grambsch, P. (2000). Modeling Survival Data: Extending the Cox Model. New York: Springer.
Turkheimer, F. E., Edison, P., Pavese, N., Roncaroli, F., Anderson, A. N., Hammers, A., . . . Brooks, D. J. (2007). Reference and Target Region Modeling of [11C]-(R)-PK11195 Brain Studies. The Journal of nuclear medicine (1978), 48(1), 158-167.
Ulrich, J. D., Ulland, T. K., Mahan, T. E., Nyström, S., Nilsson, K. P., Song, W. M., . . . Holtzman, D. M. (2018). ApoE facilitates the microglial response to amyloid plaque pathology. Journal of Experimental Medicine, 215(4), 1047-1058. doi:10.1084/jem.20171265
Verghese, P. B., Castellano, J. M., Garai, K., Wang, Y., Jiang, H., Shah, A., . . . Holtzman, D. M. (2013). ApoE influences amyloid-β (Aβ) clearance despite minimal apoE/Aβ association in physiological conditions. Proc Natl Acad Sci U S A, 110(19), E1807-1816. doi:10.1073/pnas.1220484110
Vivash, L., & O'Brien, T. J. (2016). Imaging Microglial Activation with TSPO PET: Lighting Up Neurologic Diseases? Journal of Nuclear Medicine, 57(2), 165-168. doi:10.2967/jnumed.114.141713
Wechsler, D. (1981). Wechsler Adult Intelligence Scale - Revised, Manual. New York: The Psychological Corporation.
Wechsler, D. (1987). Wechsler Memory Scale - Revised: Man ual. San Antonio: The Psychological Corporation.
Wightman, D. P., Jansen, I. E., Savage, J. E., Shadrin, A. A., Bahrami, S., Holland, D., . . . Posthuma, D. (2021). A genome-wide association study with 1,126,563 individuals identifies new risk loci for Alzheimer’s disease. Nature Genetics, 53(9), 1276-1282. doi:10.1038/s41588-021-00921-z
Wightman, D. P., Jansen, I. E., Savage, J. E., Shadrin, A. A., Bahrami, S., Holland, D., . . . Posthuma, D. (2021). A genome-wide association study with 1,126,563 individuals identifies new risk loci for Alzheimer's disease. Nat Genet, 53(9), 1276-1282. doi:10.1038/s41588-021-00921-z
Yaqub, M., Van Berckel, B. N., Schuitemaker, A., Hinz, R., Turkheimer, F. E., Tomasi, G., . . . Boellaard, R. (2012). Optimization of Supervised Cluster Analysis for Extracting Reference Tissue Input Curves in (<i>R</i>)-[<sup>11</sup>C]PK11195 Brain PET Studies. Journal of Cerebral Blood Flow & Metabolism, 32(8), 1600-1608. doi:10.1038/jcbfm.2012.59
Yin, Z., Rosenzweig, N., Kleemann, K. L., Zhang, X., Brandão, W., Margeta, M. A., . . . Butovsky, O. (2023). APOE4 impairs the microglial response in Alzheimer’s disease by inducing TGFβ-mediated checkpoints. Nature Immunology, 24(11), 1839-1853. doi:10.1038/s41590-023-01627-6
Yokokura, M., Terada, T., Bunai, T., Nakaizumi, K., Takebayashi, K., Iwata, Y., . . . Ouchi, Y. (2017). Depiction of microglial activation in aging and dementia: Positron emission tomography with [11C]DPA713 versus [11C](R)PK11195. Journal of cerebral blood flow and metabolism, 37(3), 877-889. doi:10.1177/0271678X16646788
Zou, J., Tao, S., Johnson, A., Tomljanovic, Z., Polly, K., Klein, J., . . . Kreisl, W. C. (2020). Microglial activation, but not tau pathology, is independently associated with amyloid positivity and memory impairment. Neurobiology of aging, 85, 11-21. doi:10.1016/j.neurobiolaging.2019.09.019

No competing interests reported.

AlztransABISupplementarymaterial.docx

Download PDF

Reviews received at journal
05 Jan, 2026
Reviewers agreed at journal
18 Dec, 2025
Reviews received at journal
16 Dec, 2025
Reviewers agreed at journal
16 Dec, 2025
Reviewers agreed at journal
15 Dec, 2025
Reviewers invited by journal
15 Dec, 2025
Editor assigned by journal
17 Nov, 2025
Submission checks completed at journal
17 Nov, 2025
First submitted to journal
14 Nov, 2025

You are reading this latest preprint version

Genetic Validation of ABI3 p.Ser209Phe Variant and Its effects On Early Brain Pathology in Asymptomatic Elderly Individuals

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Introduction

Materials and methods

Statistical analysis

Results

Discussion

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1