Serum long-chain omega-3 and omega-6 fatty acids differentially predict brain atrophy,white matter hyperintensities,and lacunes in individuals with cerebral small vessel disease with or without Alzheimer's disease

Abstract

Background

Several circulating fatty acids (FAs) have been linked to brain atrophy in individuals with Alzheimer's disease (AD), but the relationship between FAs and AD with or without subcortical cerebral small vessel disease (SVD) has not been investigated.

Objective

To study associations between serum FAs and brain structural pathologies in a cohort of AD and non-AD patients, with and without subcortical SVD.

Methods

Serum FAs were measured in individuals with minimal SVD (n = 28), extensive SVD (n = 29), AD with minimal SVD (n = 15) and AD with extensive SVD (n = 14). Hippocampal volume, atrophy, lacunes, and white matter hyperintensities were measured via 3.0T MRI.

Results

Higher serum linoleic acid (LA, C18:2n-6) was associated with lower periventricular lacune volumes in control individuals with minimal SVD. In individuals with AD and extensive SVD, serum docosahexaenoic acid (DHA, C22:6n-3) and the omega-3 index were associated with greater hippocampal volume.

Conclusions

This study shows disease-specific associations between serum omega-3 and omega-6 fatty acid type and brain structural features. Specifically, DHA was associated with greater hippocampal volume in those with AD and SVD co-pathology, whereas LA was associated with less periventricular lacunes in normal controls.

Keywords

Alzheimer's disease arachidonic acid cerebral small vessel disease docosahexaenoic acid eicosapentaenoic acid fatty acid hippocampus lacunar infarct leukoencephalopathies linoleic acid

Introduction

Alzheimer's disease (AD) is the leading cause of dementia in the United States,¹ and the development of effective strategies for its prevention, diagnosis, and treatment is complicated by varied clinical presentations and comingling pathologies including proteinopathies (e.g., TDP-43, alpha-synuclein) and white matter hyperintensities (WMHs) characteristic of cerebral small vessel disease (SVD). AD affects approximately 11% of individuals over the age of 65 and is the fifth-leading cause of death in this age group.² Typical pathological features include the accumulation of neurofibrillary tau tangles and amyloid-β plaques in the brain, neuroinflammation, neuronal death, and cerebral atrophy.²

SVD is the most common co-pathology seen in AD, being present in approximately 40–75% of individuals with AD diagnosis.³ It is also a major risk factor for AD as evidenced by a postmortem study showing a 12-fold increased risk of dementia in individuals over the age of 90 when both WMHs and AD pathogenic markers are present.⁴ Another postmortem study demonstrated that higher grades of cerebrovascular pathology, as evidenced by cerebral atherosclerosis and arteriolosclerosis, increased the odds of AD dementia independent of the effects of AD pathology.⁵ Subcortical infarcts associated with SVD were also shown to have a greater impact on cognitive impairment when combined with AD pathology.⁶ This suggests interacting biochemical pathways that might overlap between AD and SVD pathologies.

Fatty acids (FAs) are key structural components of brain membrane phospholipids, and are known to participate in multiple signaling and inflammatory processes in the brain through their oxylipin metabolites.^7–10 In humans, both circulating and brain FAs have been shown to be associated with the risk of acquiring AD,¹¹ brain atrophy,¹² degree of cognitive impairment,¹² and tau and amyloid-β accumulation.^12,13 For instance, a postmortem metabolic analysis of brain tissue revealed strong associations between regional levels of unsaturated fatty acids, including omega-6 linoleic acid (LA, C18:2n-6), omega-9 oleic acid (C18:1n-9) and omega-3 alpha-linolenic acid (ALA, C18:3n-3), eicosapentaenoic acid (EPA, C20:5n-3), and docosahexaenoic acid (DHA, C22:6n-3), and the severity of AD pathology and cognitive impairment.¹³ Another study reported an inverse relationship between both circulating total omega-3 and DHA percent composition and the risk of incident AD.¹¹ In one longitudinal study, a higher plasma omega-3 index (total combined percentage of plasma EPA and DHA) was positively associated with lower risk of dementia, lower rate of cognitive decline, and less atrophy of the medial temporal lobe over a 17-year period in individuals without baseline dementia.¹⁴

Several studies reported a link between circulating FAs and WMHs. One study reported a positive association between higher circulating monounsaturated FA concentrations and WMH volumes, but identified an inverse association between circulating unsaturated FAs and WMH volumes.¹⁵ In dementia-free individuals, a higher erythrocyte DHA percent composition and omega-3 index was associated with greater hippocampal volume and reduced WMH burden.¹⁶ Another study found that a lower ratio of plasma concentration of DHA to AA was associated with small ischemic lesions in the cerebrum of patients with established SVD.¹⁷ In contrast, when looking at the dietary consumption of FAs, no correlation was found between self-reported DHA intake and WMHs, suggesting that blood FA levels may better inform on dietary intake patterns and potential WMH associations.¹⁸

To date, all studies have explored the link between FAs and AD or SVD in separate cohorts where AD and vascular co-pathologies have not been well-characterized. To our knowledge, there are no comparative studies examining the role of circulating FAs on brain structural features (e.g., atrophy or WMHs) in individuals with SVD, AD, or AD with SVD co-pathology. Thus, in the present exploratory study, we measured serum FA concentrations in a cohort of AD and non-AD patients, with and without subcortical SVD as evidenced by WMH burden, to test the hypothesis that specific FAs are differentially associated with brain structural pathologies including WMHs, hippocampal volume, and whole brain atrophy.¹⁹ This study also investigated the unknown relationship between serum FAs and lacunes, another neuroimaging feature of SVD.^20,21

Methods

Participants

FA analysis was performed in serum of study participants enrolled in the Sunnybrook Vascular Brain Health (SVBH) study, a cross-sectional study designed to investigate the relationships between SVD and AD.²² Older adult study participants with and without AD were recruited into four groups of extensive or minimal WMHs seen on 3.0T MRI from 2007 to 2012. The four groups of participants were composed of 28 normal controls (NCs) with no AD and minimal WMHs, 15 individuals with AD and minimal WMHs, 29 individuals with SVD (extensive WMHs) but no AD, and 14 individuals with both AD and SVD. As previously reported, minimal WMH burden was defined as < 3 focal hyperintensities ≤ 3 mm in diameter.¹⁹ Individuals with SVD had periventricular WMHs (pWMHs) extending ≥ 5 mm from the ventricular border consistent with the Fazekas criteria,²³ and were recruited within 3 months of a transient ischemic attack without residual physical symptoms. AD was clinically diagnosed following the McKhann criteria.²⁴ The exclusion criteria consisted of participants with other neurologic or psychiatric disorders and those with cortical infarcts or cortical hyperintensities (defined as > 3 hyperintense cortical foci or any cortical lesion >3 mm in diameter) on 3.0T MRI.²²

All 86 participants were fluent in English and scored >19 on standardized Mini Mental State Examination. Clinical data such as concomitant metabolic and vascular diseases (i.e., hypertension, hyperlipidemia, coronary artery disease, and type 2 diabetes) were obtained via chart review. The Sunnybrook Research Ethics Board approved the patient study protocol, and written informed consent was obtained from all study participants.

Blood collection

Phlebotomy was performed for blood sample collection at a convenient time of day without prior fasting. Blood was collected into SST Vacutainer^™ tubes and centrifuged at 10,000 g. The serum was separated and stored at −80°C until the time of analysis, in October of 2024 (12 to 17 years after sample collection).

Magnetic resonance imaging

A Discovery MR750 (GE Healthcare) 3.0T MRI scanner was used to acquire brain scans using previously published acquisition parameters.²² The Lesion Explorer and Semi-Automatic Brain Region Extraction (sabre.brainlab.ca) imaging pipelines were used to generate regional brain volumes. The neuroimaging pipelines have been extensively validated for use in heterogeneous clinical populations with significant neurovascular and neurodegenerative pathologies.^19,25–28 In brief, after implementing standard inhomogeneity correction procedures and skull-stripping to remove non-brain tissue, the T1-weighted, proton density, T2-weighted, and T2-fluid attenuated inversion recovery (FLAIR) images were co-registered and segmented to quantify volumes of: normal appearing white matter, normal appearing grey matter, ventricular and sulcal cerebrospinal fluid, lacunes and WMHs.²⁶ Lacunes and WMHs were further segmented in relation to their proximity to the ventricles: subcortical deep WMHs (dWMHs), periventricular WMHs (pWMHs), subcortical deep lacunes (dLCN), and periventricular lacunes (pLCN). The hippocampus was segmented using the SunnyBrook Hippocampal Volumetry (SBHV) Tool and hippocampal volume was calculated as the sum of normal appearing grey matter plus normal appearing white matter.^19,29 Atrophy was measured as the brain parenchymal fraction (BPF) calculated by dividing the sum of grey and white matter volumes, by the total intracranial volume.³⁰ WMH volumes were excluded from BPF calculations.¹⁹

Fatty acid measurements

Serum FAs were analyzed in a blinded and randomized manner using gas chromatography with flame ionization detection. Lipids were extracted from 400 µL of serum aliquots using the Folch method.³¹ In brief, total lipids were extracted in 3 mL of 2:1 chloroform/methanol (Fisher Scientific, C607-4, 81 Waltham MA; Fisher Scientific, A454-4, Waltham MA) and 0.35 mL of 0.9% KCl (Fisher Scientific, P217-500, 81, Waltham, MA) containing 1 mM Na2EDTA (Sigma-Aldrich, E5134-50 g, Burlington, MA). Samples were centrifuged using a Sorvall RT 6000D tabletop centrifuge (Dupont, 83054-1, Wilmington, DE) at 2000 rpm at 0°C for 10 min, and the lower chloroform phase was transferred to a new test-tube. Two mL of chloroform were added to the original tube, which was centrifuged again at 1800 rpm and 0°C for 10 min. The bottom chloroform extract was pooled with the first, dried under nitrogen (UCDBuy, 13543-100, Davis CA) and reconstituted in 1 mL of 2:1 chloroform/isopropanol (Fisher Scientific, A464, Waltham MA). A portion of the total lipid extract (205 µL) was spiked with 0.2 mg of 1,2-diheptadecanoyl-sn-glycero-3-phosphocholine (di17:0PC) and 0.125 mg of 5-alpha cholestane internal standards (Avanti Research, 850369P-25 mg, Alabaster, AL; Sigma-Aldrich, C8003-100 MG, Burlington, MA). Samples were dried under nitrogen, reconstituted in 0.4 mL toluene and 3 mL methanol, and derivatized by adding 0.6 mL of HCl/Methanol (8:92, v/v/; Sigma-Aldrich, 320331, Burlington, MA) and incubating the test-tube at 90°C for 60 min. The reaction was stopped by adding 1 mL distilled water and the resulting fatty acid methyl esters were extracted with 1 mL hexane (Fisher Scientific, H303-4, Waltham, MA). The hexane extracts (∼900 µL) were transferred to a 1.5 mL centrifuge tube containing 450 µL of distilled water, vortexed, centrifuged at 15,871 g for 1 min and the top hexane layer was separated and reconstituted in 100 µL hexane. Individual fatty acid methyl esters were analyzed by gas chromatography with flame ionization detection (PerkinElmer Clarus 500 system, Springfield, IL) on a DB-FFAP column (30 m*0.25 mm inner diameter, 0.25 µm film thickness) using the following parameters: oven temperature set inititally at 80°C for 2 min, 10°C/min ramp to 185°C, 6°C/min ramp to 249°C and hold for 42 min for a total run time of 65 min. The injector temperature was 285°C, the detector temperature was set at 300°C, and the carrier gas (helium) flow rate was 1.3 mL/min. The injection volume per sample was 1 µL. Analytical fatty acid methyl ester standards were purchased from Nuchek Prep, and individually analyzed on the same gas chromatography system, along with a cholesterol standard (Sigma-Aldrich, C8667-500MG, Burlington MA).

Serum omega-3 index

The omega-3 index measures the sum of EPA and DHA as a percentage of total identified FAs.³² It is frequently applied to erythrocyte FAs, but given the high correlation between serum FAs and erythrocyte FAs,³³ the serum omega-3 index was calculated.

Statistical analysis

Statistical analysis was performed using either GraphPad Prism (version 9.4) or SPSS (version 23). One-way analysis of variance (ANOVA) and Bonferroni correction were performed to test for differences in age and brain structural features across the groups.¹⁹ Fisher's Exact Test was used to test for differences in gender, hypertension, hyperlipidemia, type 2 diabetes mellitus, and PUFA supplementation use across the groups. Fatty acid data were expressed as a percentage of total detected fatty acids or absolute concentrations (mg/mL). Concentration data were log-transformed prior to statistical analysis as they were not normally distributed. One-way ANOVA followed by Bonferroni's post-hoc test was performed to evaluate FA differences across the 4 groups (NC, SVD, AD and AD + SVD). Pearson's correlation was used to evaluate the relationship between serum FAs (percent composition or log-transformed concentration) and brain structural features (BPF, hippocampal volume, total WMHs, dWMHs, pWMHs, dLCN, and pLCN) in each group. Pearson's correlation was also used to evaluate the relationship between age and FAs. The relationship between gender and FAs were evaluated using Welch's T-Tests. False discovery rate (FDR) correction was applied to all data at a Q of 10% (n = 16 detected FAs).

Results

Participant characteristics are shown in Table 1. As previously reported,^19,34 age in years ranged from 59 to 87 (mean = 72), 43 to 89 (mean = 72), 63 to 88 (mean = 73), and 61 to 81 (mean = 80) in the NC, SVD, AD, and AD + SVD groups respectively. The AD + SVD group was significantly older than the SVD group (p = 0.046 by one-way ANOVA and Bonferroni's post-hoc test; mean difference = 7.89 years). Participants were 44%, 48%, 60%, and 43% male in the NC, SVD, AD, and AD + SVD groups respectively. Hypertension was most common in the SVD (72%, n = 25) and the AD + SVD (62%, n = 13) groups, compared to the AD group (33%, n = 15) and NC group (40%, n = 25). Hyperlipidemia was also more prevalent in SVD (52%, n = 25), AD (47%, n = 15), and AD + SVD (46%, n = 13), than in NCs (29%, n = 24). Type 2 diabetes mellitus was most prevalent in the AD + SVD (15%, n = 13) and SVD (13%, n = 24) groups, compared to the AD group (7%, n = 15) and NCs (0%, n = 24). NCs were more likely to use PUFA supplements (11%, n = 25), followed by the SVD group (3%, n = 29), whereas no subjects in the AD or AD + SVD groups used PUFA supplements.

Table 1.

Participant characteristics (Adapted from Otoki et al.¹⁹ and Yu et al.³⁴).

	NC				SVD				AD				AD + SVD
	Mean		SD	n	Mean		SD	n	Mean		SD	N	Mean		SD	n	p
Demographics
Age (y)	72.0	±	7.9	28	71.8	±	11.0	29	73.0	±	7.2	15	79.7	±	6.8	14	0.0401
Gender (% male)	44%			28	48%			29	60%			15	43%			14	0.7902
Vascular Factors
Hypertension	40%			25	72%			25	31%			15	62%			13	0.0468
Hyperlipidemia	29%			24	52%			25	44%			15	46%			13	0.4088
Type 2 diabetes mellitus	0%			24	13%			25	6%			15	15%			13	0.1914
PUFA supplements	12%			25	3%			29	0%			15	0%			14	0.4818
Brain Structural Features
Brain parenchymal fraction	0.78	±	0.05	28	0.76	±	0.05	28	0.75	±	0.04	14	0.72	±	0.03	13	0.0038
Total WMH (mm³)	5136.8	±	4951.2	23	31628.54	±	21613.3	28	3555.1	±	2059.9	12	32553.4	±	22728.4	13	<0.0001
dWMH (mm³)	674.7	±	700.6	28	2552.8	±	2493.1	28	787.0	±	814.0	13	3135.7	±	2231.7	13	<0.0001
pWMH (mm³)	5087.7	±	6129.7	28	29020.9	±	21550.9	28	3208.7	±	2281.2	13	29280.9	±	20979.0	13	<0.0001
Hippocampal volume (%)	0.52	±	0.08	23	0.49	±	0.07	24	0.39	±	0.07	11	0.41	±	0.09	11	<0.0001
dLCN (mm³)	9.56	±	17.31	27	21.73	±	35.52	26	9.54	±	13.11	11	68.15	±	119.63	13	0.0197
pLCN (mm³)	116.48	±	208.21	27	994.62	±	1257.21	26	57.69	±	100.94	12	1243.23	±	1472.27	13	<0.0001

Participant characteristics of the normal control (NC), small vessel disease (SVD), Alzheimer's disease (AD) and AD + SVD groups. Continuous data (age, BPF, total WMH, dWMH, pWMH, hippocampal volume, dLCN, and pLCN) are presented as mean ± SD. Non-continuous data (gender, hypertension, hyperlipidemia, type 2 diabetes mellitus, and PUFA supplementation) are presented as a percentage of the total participants in each group. BPF was used to measure atrophy as described in the Methods section. Continuous data were analyzed by one-way analysis of variance (ANOVA) and corrected using Bonferroni's multiple comparisons. Non-continuous data were analyzed using Fisher's Exact Test. Sample size variations are due to missing data from one or a few subjects per group.

Structural features

Brain structural features, which have been previously reported in this cohort, are presented in Table 1.^19,34 AD + SVD participants had significantly lower BPF (i.e., more atrophy) compared to controls (p_ANOVA < 0.0038). Hippocampal volume (reported as % hippocampal volume of total intracranial volume) was significantly lower in the AD and AD + SVD groups compared to controls and the SVD group (p_ANOVA < 0.0001). A representative MRI demonstrating hippocampal volume in each of the groups is shown in Figure 1.

Figure 1.

T1-weighted MRI axial images highlighting hippocampal volume in a cognitively normal older adult with minimal WMH burden (NC), an individual diagnosed with AD with minimal WMH burden (AD), an individual diagnosed with SVD, and an individual with mixed AD and SVD (AD + SVD). The right and left hippocampi are highlighted in red and green respectively with corresponding three-dimensional renderings below each axial image.

Total WMHs, dWMHs, and pWMHs were significantly higher in the SVD and AD + SVD groups than in NCs and the AD group (p_ANOVA < 0.0001). The volume of dLCN was greater in AD + SVD participants than in NCs (p_ANOVA = 0.0197). The AD + SVD and SVD groups also demonstrated greater volume of pLCN than the NC or AD groups (p_ANOVA <0.0001).

Fatty acid measurements

FA measurements are shown in Table 2 as percent composition. C8:0, C10:0, and C14:1 were not detected. C18:1 n-9 cis and C18:1n-7 were measured as a combined value due to peak coelution. FA concentrations (mg/mL) are shown in Supplemental Table 1. One-way ANOVA revealed no significant differences in serum FAs (percent composition and absolute concentrations) between the groups.

Table 2.

Fatty acid percent composition.

		NC				SVD				AD				AD + SVD
Fatty Acid	Common Name	Mean		SD	n	Mean		SD	n	Mean		SD	n	Mean		SD	n
C12:0 (%)	Lauric acid	0.21	±	0.15	28	0.29	±	0.24	29	0.25	±	0.29	15	0.21	±	0.13	14
C13:0 (%)		3.61	±	1.54	28	3.89	±	1.94	29	3.25	±	1.37	15	3.75	±	2.04	14
C14:0 (%)	Myristic acid	1.68	±	0.79	28	1.99	±	1.19	29	1.81	±	1.38	15	1.61	±	0.53	14
C15:0 (%)		0.32	±	0.10	28	0.36	±	0.14	29	0.33	±	0.14	14	0.33	±	0.09	14
C16:0 (%)	Palmitic acid	28.25	±	4.40	28	30.13	±	6.82	29	29.11	±	5.95	15	28.88	±	4.45	14
C16:1 (%)	Palmitoleic acid	1.39	±	0.61	28	1.28	±	0.59	29	1.33	±	0.43	15	1.23	±	0.44	14
C18:0 (%)	Stearic acid	9.48	±	1.68	28	10.80	±	2.77	29	9.83	±	1.39	15	10.23	±	1.76	14
C18:1 n-9 cis (%)plus C18:1 n-7 (%)	Oleic acidplusVaccenic acid	18.87	±	6.21	28	18.25	±	5.88	29	18.43	±	3.78	15	17.86	±	4.18	14
C18:2n-6 (%)	Linoleic acid (LA)	22.35	±	5.95	28	19.97	±	7.44	29	21.98	±	6.13	15	21.61	±	4.49	14
C18:3 n-6 (%)	γ-Linolenic acid	0.67	±	0.23	27	0.67	±	0.24	28	0.70	±	0.12	15	0.74	±	0.18	14
C18:3n-3 (%)	α-Linolenic acid (ALA)	0.64	±	0.36	26	0.66	±	0.45	27	0.54	±	0.21	15	0.57	±	0.23	14
C20:4n-6 (%)	Arachidonic acid (AA)	10.15	±	4.68	28	9.62	±	3.79	29	9.97	±	2.49	15	10.45	±	3.05	14
C20:5n-3 (%)	Eicosapentaenoic acid (EPA)	0.42	±	0.59	27	0.43	±	0.33	29	0.29	±	0.15	15	0.57	±	0.55	12
C22:5 n-3 (%)	Docosapentaenoic acid	0.54	±	0.19	27	0.52	±	0.12	29	0.65	±	0.32	15	0.56	±	0.11	14
C22:6n-3 (%)	Docosahexaenoic acid (DHA)	1.40	±	0.75	28	1.15	±	0.53	29	1.54	±	1.00	15	1.38	±	0.77	14
Omega-3 Index		1.82	±	1.15	28	1.58	±	0.69	29	1.83	±	0.94	15	1.95	±	1.20	14

Fatty acid % composition and the omega-3 index in the normal control (NC), small vessel disease (SVD), Alzheimer's disease (AD) and AD + SVD groups. There was no significant difference between the means of the groups using one-way analysis of variance. P values were considered significant if they were below the typical 0.05 threshold for significance. Group numbers may vary if a FA was not detected in the sample.

Correlation between FAs and WMH volumes and lacunes

Supplemental Table 2 shows the correlations between percent FA composition and structural features including WMHs, lacunes, BPF (i.e., atrophy) and hippocampal volumes within each of the control, SVD, AD and AD + SVD groups. As shown, several FAs were correlated with WMHs and lacunes, but only the inverse correlation between LA % and pLCN in NCs (Figure 2, r = -0.5136, p = 0.0061) remained significant after FDR correction. Supplemental Table 3 shows the same correlations but for the absolute FA concentration data. FDR correction led to no significant correlations between the absolute concentration of FAs and structural features within any of the four groups.

Figure 2.

Scatter plot of a) Serum omega-3 index and hippocampal volume in AD + SVD (n = 11), b) DHA % and hippocampal volume in AD + SVD (n = 11), and c) LA % and periventricular lacunes (pLCN) in NC (n = 27). Variations in sample size are due to missing MRI data from a few subjects per group. Data was analyzed via Pearson's Correlation. DHA % was positively correlated with hippocampal volume (r = 0.7939, p = 0.0035) in AD + SVD. The omega-3 index was positively correlated with hippocampal volume in AD + SVD (r = 0.7595, p = 0.0067). LA % was negatively correlated with the volume (mm³) of pLCN (r = -0.5136, p = 0.0061) in NCs. P values were considered significant if they were below the typical 0.05 threshold for significance, and were below the FDR-corrected Q value.

Correlations between FAs and hippocampal volume

After FDR correction, DHA percent composition was positively correlated with hippocampal volume but only in the AD + SVD group (Figure 2, r = 0.786, p = 0.0041). Similarly, the omega-3 index was positively correlated with hippocampal volume in the AD + SVD group (Figure 2, r = 0.7595, p = 0.0067). Other FAs were correlated with hippocampal volume in the AD and SVD groups, but did not maintain significance after FDR correction (Supplemental Table 2). To determine whether it was worth including age and gender as potential confounders to the model, we tested whether DHA or the omega-3 index were modified by either variables. In the overall cohort, Pearson's correlations revealed no significant associations between age and either the omega-3 index (p = 0.8149, r = 0.026) or DHA (p = 0.590, r = 0.026). Likewise, no significant differences were observed between males and females in the omega-3 index (p = 0.3614, t(77.36) = 0.918) or DHA (p = 0.2618, t(68.51) = 1.131), as determined by Welch's T-Tests.

Correlation between FAs and BPF

No significant correlations were observed between FAs and BPF after FDR correction.

Discussion

In the present exploratory study, we found disease-specific associations between serum omega-3 fatty acids and hippocampal volume in individuals with AD + SVD, and omega-6 fatty acids and periventricular lacunes in controls with minimal SVD. In the AD + SVD group, DHA percent composition and the omega-3 index were positively associated with hippocampal volume. In NCs, a higher percent serum composition of LA was associated with a lower volume of pLCN. Our findings point to a possible protective effect of DHA in individuals with both SVD and AD co-pathology, and a potential protective role of LA in NCs that appears to be lost with the onset of AD and/or SVD.

DHA was associated with greater hippocampal volume in the AD + SVD group, suggesting likely beneficial effects on brain structural integrity in individuals with both AD and SVD co-pathology. DHA is an essential structural component of neuronal, glial, and endothelial membrane phospholipids in gray matter and synapses.³⁵ Our findings suggest a structural role of DHA in maintaining hippocampus integrity in the AD + SVD group, which had significant reductions in hippocampal volume compared to the other groups (Table 1). DHA is also a precursor to lipid mediators (e.g., resolvins) that are involved in neurogenesis and inflammation resolution.³⁵ The observed increase in hippocampal volume in AD + SVD patients with higher levels of DHA is consistent with prior research demonstrating that DHA is protective against age-related reductions in cortical thickness and loss of white and grey matter volumes.^18,36 Our observations are further supported by another study reporting lower levels of DHA in the hippocampus of AD patients compared to healthy controls, although AD with and without SVD was not differentiated.³⁷ Notably, the AD + SVD group was significantly older than the SVD group, which may have impacted the observed associations between DHA with hippocampal volume. However, across all groups, there was no significant relationship between age and DHA, suggesting that the relationship between DHA and hippocampal volume may be independent of age. Given our sample size, our study should be seen as exploratory and future studies in larger cohorts are needed to confirm our findings.

In our study, a high omega-3 index was also associated with greater hippocampal volume in the AD + SVD group, which was likely mediated by DHA. Higher serum omega-3 PUFA levels have previously been linked to better white matter integrity,³⁸ less atrophy,³⁹ and larger hippocampal volumes.^16,40 These findings contrast with one study that reported greater total hippocampal volume in AD individuals with a low erythrocyte omega-3 index.¹² This could be due to a failure to tease apart individuals with AD alone from individuals with both AD and SVD co-pathologies. We demonstrated no association between DHA or the omega-3 index and any structural features in the AD group alone. These findings suggest that DHA may play a more significant role in comingling AD and SVD pathologies.

LA was associated with less pLCN in controls suggesting protective effects on cerebrovascular disease pathology. There is very little LA in the brain (<1%) because it rapidly gets beta-oxidized or converted into pro-inflammatory lipid mediators upon entering the brain.^41,42 Thus, while these observations are unexpected, the possible neuroprotective effects of LA are an emerging trend in the literature. In the same cohort, we recently reported that LA bound to phospholipids is associated with less atrophy in NCs.¹⁹ This is consistent with another study which reported that lower baseline levels of the LA-derived oxylipin, 9-hydroxyoctadecadienoic acid, predicted white matter damage in patients with relapsing remitting multiple sclerosis.⁴³ In the brain, LA is also recycled into cholesterol and other structural lipids that may be important in neurodegenerative diseases.⁴⁴ LA and its oxylipin metabolites have also been linked to inflammatory vascular repair mechanisms that influence endothelial cell nitric oxide production, migration, and proliferation.^45–47 Endothelial cell dysfunction is a key driver of impaired cerebral perfusion and can result in the lacunes and WMHs seen in SVD.^48,49 Our findings suggest that LA may be protective against brain structural abnormalities (e.g., lacunes) prior to the onset of AD or SVD pathologies, as these effects were lost in the AD, SVD and AD + SVD groups. Future studies should investigate whether LA could delay AD or SVD disease onset in healthy elderly.

The lack of significant FA associations in the SVD group is surprising considering prior literature showing a relationship between FAs (e.g., monounsaturated fatty acids) and WMHs.¹⁵ These associations, however, are better established in dementia free individuals.^16,50 It is possible this association is lost once there is progression to SVD with memory impairments. Thus, FAs may be more predictive of future risk or advancement of SVD cognitive symptoms rather than WMH burden in established SVD.¹² Longitudinal follow-up with cognitive testing is needed, as it is possible such a relationship would be demonstrated over time, or the relationship (between FAs and WMHs) was missed due to our small sample size.

As this study was an exploratory post-hoc analysis, the small sample sizes of the groups presented a limitation. Additionally, our patient population was non-fasting; however, this was accounted for by using percent values to demonstrate relative FA abundance. Findings were strengthened by analyzing both FA percent composition and absolute concentration which exhibited parallel trends in many cases (Supplemental Tables 2 and 3). Our study was also strengthened by separating AD and SVD from AD + SVD participants using WMHs as seen on MRI. This allowed us to evaluate group-specific correlations between FAs and structural changes and account for coexisting pathologies.

In conclusion, this study demonstrated group-specific relationships between serum FAs and structural neuropathologic features of AD and SVD. Our work suggests that altered serum FA levels (particularly omega-3 DHA and omega-6 LA), may play important roles in the pathogenesis of hippocampal atrophy, and lacunes that are thought to precede cognitive decline.^51–53 DHA appears to play a protective role in maintaining hippocampal integrity in individuals with comingling SVD and AD pathology, and LA may have an important role in preventing brain lacunes in healthy elderly. Additional research is needed to confirm our findings in a larger cohort and to further explore the mechanisms behind these relationships.

Supplemental Material

sj-docx-1-alz-10.1177_13872877251401572 - Supplemental material

Supplemental material, sj-docx-1-alz-10.1177_13872877251401572

Footnotes

Acknowledgements

We extend our gratitude to the study participants for their time and contribution to our study.

ORCID iD

Ameer Y Taha

Ethical considerations

This study was approved by the Sunnybrook Research Ethics Board (REB 063-2006) on March 24, 2006. This research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki.

Consent to participate

All participants provided written informed consent prior to participating. Written informed consent was obtained from a legally authorized representative for anonymized patient information to be published in this article.

Consent for publication

Not applicable

Author contribution(s)

Leslie M Yedor: Conceptualization; Formal analysis; Investigation; Visualization; Writing – original draft; Writing – review & editing.

Kathryn O Chambers: Investigation; Writing – review & editing.

Maria Moran-Garrido: Investigation; Writing – review & editing.

Sofia Perfetto: Data curation; Writing – review & editing.

Willian Z Lin: Data curation; Writing – review & editing.

Si Won Ryoo: Data curation; Writing – review & editing.

Celine Huang: Data curation; Writing – review & editing.

Joel Ramirez: Data curation; Formal analysis; Visualization; Writing – review & editing.

Demetrios J Sahlas: Data curation; Writing – review & editing.

Fuqiang Gao: Data curation; Writing – review & editing.

Mario Masellis: Data curation; Writing – review & editing.

Richard H Swartz: Data curation; Writing – original draft.

Jacqueline A Pettersen: Data curation; Writing – review & editing.

Coral Barbas: Supervision; Writing – review & editing.

Sandra E Black: Data curation; Writing – review & editing.

Walter Swardfager: Funding acquisition; Supervision; Writing – review & editing.

Ameer Y Taha: Conceptualization; Funding acquisition; Project administration; Supervision; Writing – review & editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding for this study was provided by the Alzheimer's Association (2018-AARGD-591676) and the USDA National Institute of Food and Agriculture (Hatch Project #1008787) to Ameer Y. Taha. Walter Swardfager received financial support from a Team Grant from the Canadian Institutes of Health Research (CIHR) Institutes of Aging (176444); The Natural Sciences and Engineering Research Council of Canada (RGPIN- 2017-06962), the Alzheimer's Association & Brain Canada (AARG501466), Weston Brain Institute & Alzheimer's Research UK, Alzheimer's Association and Michael J. Fox Foundation (BAND3), and the Canada Research Chairs Program (CRC-2020- 00353). The funders were not involved in any aspects of the study.

Declaration of conflicting interests

The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The data supporting the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Supplemental material

Supplemental material for this article is available online.

References

Iadecola

Duering

Hachinski

, et al. Vascular cognitive impairment and dementia: JACC scientific expert panel. J Am Coll Cardiol 2019; 73: 3326–3344.

Alzheimer's Association. 2024 Alzheimer's disease facts and figures. Alzheimers Dement 2024; 20: 3708–3821.

Kapasi

DeCarli

Schneider

. Impact of multiple pathologies on the threshold for clinically overt dementia. Acta Neuropathol 2017; 134: 171–186.

Kawas

Kim

Sonnen

, et al. Multiple pathologies are common and related to dementia in the oldest-old: the 90+ study. Neurology 2015; 85: 535–542.

Arvanitakis

Capuano

Leurgans

, et al. Relation of cerebral vessel disease to Alzheimer's disease dementia and cognitive function in elderly people: a cross-sectional study. Lancet Neurol 2016; 15: 934–943.

Schneider

Boyle

Arvanitakis

, et al. Subcortical infarcts, Alzheimer's disease pathology, and memory function in older persons. Ann Neurol 2007; 62: 59–66.

de Carvalho

Caramujo

. The various roles of fatty acids. Molecules 2018; 23: 2583.

Belkind-Gerson

Carreón-Rodríguez

Contreras-Ochoa

, et al. Fatty acids and neurodevelopment. J Pediatr Gastroenterol Nutr 2008; 47: S7–S9.

Taha

Cheon

, et al. Altered fatty acid concentrations in prefrontal cortex of schizophrenic patients. J Psychiatr Res 2013; 47: 636–643.

10.

Ebright

Duro

Chen

, et al. Effects of APOE4 on omega-3 brain metabolism across the lifespan. Trends Endocrinol Metab 2024; 35: 745–757.

11.

Sala-Vila

Tintle

Westra

, et al. Plasma omega-3 fatty acids and risk for incident dementia in the UK Biobank Study: a closer look. Nutrients 2023; 15: 4896.

12.

Rouch

Virecoulon Giudici

Cantet

, et al. Associations of erythrocyte omega-3 fatty acids with cognition, brain imaging and biomarkers in the Alzheimer's disease neuroimaging initiative: cross-sectional and longitudinal retrospective analyses. Am J Clin Nutr 2022; 116: 1492–1506.

13.

Snowden

Ebshiana

Hye

, et al. Association between fatty acid metabolism in the brain and Alzheimer disease neuropathology and cognitive performance: a nontargeted metabolomic study. PLoS Med 2017; 14: e1002266.

14.

Thomas

Baillet

Proust-Lima

, et al. Blood polyunsaturated omega-3 fatty acids, brain atrophy, cognitive decline, and dementia risk. Alzheimers Dement 2021; 17: 407–416.

15.

Sun

Guo

, et al. Associations of circulating metabolites with cerebral white matter hyperintensities. J Neurochem 2023; 166: 414–423.

16.

Satizabal

Himali

Beiser

, et al. Association of red blood cell omega-3 fatty acids with MRI markers and cognitive function in midlife: the Framingham Heart Study. Neurology 2022; 99: e2572–e2582.

17.

Okumura

Sato

Takahashi

, et al. Small ischemic lesions accompanying intracerebral hemorrhage: the underlying influence of old lacunes and polyunsaturated fatty acids. Nutr Metab Cardiovasc Dis 2024; 34: 1157–1165.

18.

Sala-Vila

Arenaza-Urquijo

Sánchez-Benavides

, et al. DHA Intake relates to better cerebrovascular and neurodegeneration neuroimaging phenotypes in middle-aged adults at increased genetic risk of Alzheimer disease. Am J Clin Nutr 2021; 113: 1627–1635.

19.

Otoki

Shen

, et al. Quantitative lipidomic analysis of serum phospholipids reveals dissociable markers of Alzheimer's disease and subcortical cerebrovascular disease. J Alzheimers Dis 2023; 93: 665–682.

20.

Duering

Csanadi

Gesierich

, et al. Incident lacunes preferentially localize to the edge of white matter hyperintensities: insights into the pathophysiology of cerebral small vessel disease. Brain 2013; 136: 2717–2726.

21.

Kwon

Cho

Lee

, et al. Evolution of acute lacunar lesions in terms of size and shape: a PICASSO sub-study. J Neurol 2019; 266: 766–772.

22.

Swardfager

Scola

, et al. Peripheral lipid oxidative stress markers are related to vascular risk factors and subcortical small vessel disease. Neurobiol Aging 2017; 59: 91–97.

23.

Fazekas

Chawluk

Alavi

, et al. MR Signal abnormalities at 1.5T in Alzheimer's dementia and normal aging. AJR Am J Roentgenol 1987; 149: 351–356.

24.

McKhann

Knopman

Chertkow

, et al. The diagnosis of dementia due to Alzheimer's disease: recommendations from the national institute on aging-Alzheimer's association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement 2011; 7: 263–269.

25.

Gibson

Gao

Black

, et al. Automatic segmentation of white matter hyperintensities in the elderly using FLAIR images at 3T. J Magn Reson Imaging 2010; 31: 1311–1322.

26.

Ramirez

Gibson

Quddus

, et al. Lesion explorer: a comprehensive segmentation and parcellation package to obtain regional volumetrics for subcortical hyperintensities and intracranial tissue. Neuroimage 2011; 54: 963–973.

27.

Ramirez

Scott

Black

. A short-term scan-rescan reliability test measuring brain tissue and subcortical hyperintensity volumetrics obtained using the lesion explorer structural MRI processing pipeline. Brain Topogr 2013; 26: 35–38.

28.

Dade

Gao

Kovacevic

, et al. Semiautomatic brain region extraction: a method of parcellating brain regions from structural magnetic resonance images. Neuroimage 2004; 22: 1492–1502.

29.

Nestor

Gibson

Gao

F-Q

, et al. A direct morphometric comparison of five labeling protocols for multi-atlas driven automatic segmentation of the hippocampus in Alzheimer's disease. Neuroimage 2013; 66: 50–70.

30.

Swardfager

Cogo-Moreira

Masellis

, et al. The effect of white matter hyperintensities on verbal memory: mediation by temporal lobe atrophy. Neurology 2018; 90: e673–e682.

31.

Taha

Metherel

Stark

. Comparative analysis of standardised and common modifications of methods for lipid extraction for the determination of fatty acids. Food Chem 2012; 134: 427–433.

32.

Harris

. The omega-3 index: clinical utility for therapeutic intervention. Curr Cardiol Rep 2010; 12: 503–508.

33.

Harris

. The omega-3 index as a risk factor for coronary heart disease1. Am J Clin Nutr 2008; 87: 1997S–2002S.

34.

Hennebelle

Sahlas

, et al. Soluble epoxide hydrolase-derived linoleic acid oxylipins in serum are associated with periventricular white matter hyperintensities and vascular cognitive impairment. Transl Stroke Res 2019; 10: 522–533.

35.

Bazinet

Layé

. Polyunsaturated fatty acids and their metabolites in brain function and disease. Nat Rev Neurosci 2014; 15: 771–785.

36.

Witte

Kerti

Hermannstädter

, et al. Long-chain omega-3 fatty acids improve brain function and structure in older adults. Cereb Cortex 2014; 24: 3059–3068.

37.

Lukiw

Cui

Marcheselli

, et al. A role for docosahexaenoic acid-derived neuroprotectin D1 in neural cell survival and Alzheimer disease. J Clin Invest 2005; 115: 2774–2783.

38.

Virtanen

Siscovick

Lemaitre

, et al. Circulating omega-3 polyunsaturated fatty acids and subclinical brain abnormalities on MRI in older adults: the cardiovascular health study. J Am Heart Assoc 2013; 2: e000305.

39.

Samieri

Maillard

Crivello

, et al. Plasma long-chain omega-3 fatty acids and atrophy of the medial temporal lobe. Neurology 2012; 79: 642–650.

40.

Pottala

Yaffe

Robinson

, et al. Higher RBC EPA + DHA corresponds with larger total brain and hippocampal volumes: WHIMS-MRI study. Neurology 2014; 82: 435–442.

41.

Taha

Hennebelle

Yang

, et al. Regulation of rat plasma and cerebral cortex oxylipin concentrations with increasing levels of dietary linoleic acid. Prostaglandins Leukot Essent Fatty Acids 2018; 138: 71–80.

42.

DeMar

Jr. Lee

, et al. Brain elongation of linoleic acid is a negligible source of the arachidonate in brain phospholipids of adult rats. Biochim Biophys Acta 2006; 1761: 1050–1059.

43.

Fung

van Lingen

Broos

, et al. 9-HODE Associates with thalamic atrophy and predicts white matter damage in multiple sclerosis. Mult Scler Relat Disord 2024; 92: 105946.

44.

Cunnane

Trotti

Ryan

. Specific linoleate deficiency in the rat does not prevent substantial carbon recycling from [(14)C]linoleate into sterols. J Lipid Res 2000; 41: 1808–1811.

45.

Taylor

Aukema

, et al. Role of oxylipins generated from dietary PUFAs in the modulation of endothelial cell function. Prostaglandins Leukot Essent Fatty Acids 2020; 160: 102160.

46.

Rodrigues

Vinolo

Sato

, et al. Oral administration of linoleic acid induces new vessel formation and improves skin wound healing in diabetic rats. PLoS One 2016; 11: e0165115.

47.

Smith

Muffley

Bell

, et al. Unsaturated fatty acids induce mesenchymal stem cells to increase secretion of angiogenic mediators. J Cell Physiol 2012; 227: 3225–3233.

48.

Quick

Moss

Rajani

, et al. A vessel for change: endothelial dysfunction in cerebral small vessel disease. Trends Neurosci 2021; 44: 289–305.

49.

de Leeuw

de Kleine

Frijns

, et al. Endothelial cell activation is associated with cerebral white matter lesions in patients with cerebrovascular disease. Ann N Y Acad Sci 2002; 977: 306–314.

50.

Harshfield

Sands

Tuladhar

, et al. Metabolomic profiling in small vessel disease identifies multiple associations with disease severity. Brain 2022; 145: 2461–2471.

51.

Mormino

Kluth

Madison

, et al. Episodic memory loss is related to hippocampal-mediated beta-amyloid deposition in elderly subjects. Brain 2009; 132: 1310–1323.

52.

Gorbach

Pudas

Lundquist

, et al. Longitudinal association between hippocampus atrophy and episodic-memory decline. Neurobiol Aging 2017; 51: 167–176.

53.

Black

Gao

Bilbao

. Understanding white matter disease: imaging-pathological correlations in vascular cognitive impairment. Stroke 2009; 40: S48–S52.

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.

0.03 MB

0.00 MB