Fatty Acid-Binding Protein 3 Is a Marker of Neurodegeneration and White Matter Hyperintensity Burden in Mexican American Older Adults

Abstract

We examined ethnoracial differences in fatty acid binding protein (FABP)—a family of intracellular lipid carriers—and clarified FABP3 associations with gray and white matter. Relative to Mexican Americans (MAs), FABP3 was higher in Non-Hispanic Whites (NHWS, p < 0.001). Regressions revealed, independent of traditional AD markers, FABP3 was associated with neurodegeneration (B = –0.08, p = 0.003) and WMH burden (B = 0.18, p = 0.03) in MAs, but not in NHWs (ps > 0.18). Findings suggest FABP3 is related to neural health within MAs and highlight its potential as a prognostic marker of brain health in ethnoracially diverse older adults.

Keywords

Alzheimer’s disease FABP3 Hispanic/Latino lipid dyshomeostasis mild cognitive impairment

INTRODUCTION

While the aggregation of amyloid and tau proteins represent key features of Alzheimer’s disease (AD), lipid dyshomeostasis plays a crucial role in the development and progression of this pathology [1]. The Apolipoprotein E (ApoE) is a key protein involved in lipid transport and metabolism and genome-wide association studies have identified the ApoE ɛ4 isoform as one of the strongest genetic risk factors for AD [2, 3]. However, most studies exploring ApoE have largely taken place in racially homogenous samples of non-Hispanic Whites (NHWs), and there is evidence to suggest that 1) ApoE ɛ4 isoform occurs in lower frequencies among individuals from non-European ancestral backgrounds and 2) the cognitive and neural consequences conferred by the ɛ4 isoform differ across ethnoracial groups [3 –5].

In addition to apolipoproteins, a growing body of literature suggests that fatty acid binding proteins (FABP)—another family of intracellular lipid carriers—play a role in the pathogenesis of neurodegenerative disorders [6 –9]. There are three main types of FABPs (i.e., FABP-3, –5, –7) within the central nervous system responsible for the intracellular transport of polyunsaturated fatty acids. Changes in FABP occurs with age and within the context of cardiometabolic conditions; these changes increase the susceptibility of fatty acids to lipid oxidation and promote amyloid-β plaque formation underlying neurodegeneration [6]. Knock-out mice studies have revealed that FABP ablation reduces phospholipid mass and alters cortical gray and white matter [10, 11]. Studies in human (predominantly NHW samples) have shown that plasma FABP generally increases across the AD continuum, and that baseline levels of FABP-3 in cognitively normal older adults have been predictive of future entorhinal atrophy irrespective of phosphorylated tau [12 –15].

Recent research has established that Hispanic/Latino (H/L) older adults have higher rates of chronic cardiometabolic health conditions (e.g., diabetes, hypertension) and display more severe small vessel ischemic disease as indicated by white matter lesion burden on neuroimaging scans [16 –18]. Furthermore, postmortem examinations of brains have revealed that H/Ls have more severe vascular and mixed, as opposed to traditional, AD pathology when compared to NHWs [19]. Given H/Ls are also disproportionately affected by AD [16], FABPs may be an important early biomarker of a lipid-driven AD cascade within this group.

Importantly, recent work in the Healthy Aging Brain Study –Health Disparities (HABS-HD) cohort identified that FABP3 is an important biomarker in blood-based proteomic profiles of mild cognitive impairment and AD in both H/Ls and NHWs [20, 21]. However, the relative contribution of FABP3 to these biomarker profiles was higher in H/Ls and notably outperformed plasma markers of amyloid and tau [21]. Given the need to better understand distinct biological pathways and ethnoracial differences in AD pathogenesis, the current study sought to extend previous findings by 1) clarifying whether there are racial/ethnic differences in FABP3 levels, and 2) examining whether FABP3, beyond traditional AD plasma markers, is associated with neuroimaging markers of gray and white matter degeneration in AD predilection sites within a large-scale community-based sample of older adults.

METHODS

Data

Data for the present study were obtained from the HABS-HD database [22]. HABS-HD (formerly the Healthy Aging Brain in Latino Elders [HABLE] study) was launched by Drs. Sid O-Bryant (Executive Director) and Leigh Johnson (Associate Director) and is a single-site study conducted at The University of North Texas Health Science Center’s Institute for Translational Research in Fort Worth, TX [23]. HABS-HD utilizes community-based participatory research approaches and is dedicated to eliminating disparities in Alzheimer’s disease. HABS-HD was approved by the Institutional Review Boards of all participating sites and written informed consent was obtained for all study participants. Enrollment criteria for the HABS-HD study are described in detail elsewhere [23].

Data for 1,705 participants were available for use and downloaded on 3/1/22. The present study consisted of 1,369 participants (693 Mexican Americans and 676 NHWs) that were 1) determined to be cognitively unimpaired or have a diagnosis of mild cognitive impairment (MCI) at their baseline study visit and 2) had the following data available: serum FABP-3 levels; plasma AD protein markers; MRI data; medical/health information (e.g., history of heart disease, diabetes); demographic information, and apolipoprotein E (APOE) genotyping.

Cognitive status

Participants completed neuropsychological testing on measures of general cognition (Mini-Mental Status Examination [MMSE], Clinical Dementia Rating Scale [CDR]), attention/executive functioning (Weschler Memory Scale: 3rd Edition (WMS-III): Digit Span; Trail Making Test Parts A and B; Digit Symbol Substitution), verbal memory (WMS-III: Logical Memory; Spanish English Verbal Learning Test), language (Phonemic and category fluency), and premorbid intellectual functioning (American Version of the National Adult Reading Test). Cognitive diagnoses were based on the following algorithmic criteria and were verified in a clinician consensus review. Cognitively Unimpaired: Clinical Dementia Rating (CDR) sum of boxes score = 0; neuropsychological test scores considered broadly within normal limits; and no complaints of cognitive change. MCI diagnosis: CDR sum of boxes score = 0.5–2; performance ≤1.5 standard deviations below z-score adjusted norms on at least one cognitive test; and complaints of cognitive change [23].

Plasma and genetic markers

Assay preparation was completed using a custom automatic StarPlus system from Hamilton Robotics. Baseline levels serum FABP3 were measured using a multi-plex biomarker assay platform using electrochemiluminescence (ECL) in accordance with previously published protocols [23, 24]. The HABS-HD Biomarker Core has reportedly performed >20,000 assays with excellent coefficient of variation ≤10% for the ECL platform [21, 23]. Baseline levels of plasma amyloid-β 40 (Aβ₄₀)/42 (Aβ₄₂) and total tau (t-tau) were assessed using the ultra-sensitive Quanterix Simoa (single molecule array) technology platform. The HABS-HD Biomarker Core has performed >5,000 assays and reported excellent coefficient of variation ≤5% for the Quanterix platform [21, 23].

APOE ɛ4 positivity was determined by the possession of at least one ɛ4 allele. Genotyping was performed with the TaqMan Genotyping Kits for rs429158 and rs7412 (ThermoFisher). The 7,500 real-time PCR System (Applied Biosystems) was used for target detection and amplification, with both positive and negative control subjects included on all runs. Hardy-Weinberg equilibrium was confirmed for all APOE genotype frequencies [21, 23].

Cardiometabolic health burden

Anthropomorphic measurements (i.e., height, weight, abdominal circumference, blood pressure readings) were collected during the baseline study visit. Each participant completed fasting blood and clinical labs including a complete blood count with differential, comprehensive metabolic and lipid panels at a Quest laboratory location [19]. Elevated waist-circumference (W-C; women >35, men >40 inches), blood pressure (systolic >129 or diastolic >84 mm Hg), triglycerides (>149 mg/dL), glucose (>100 mg/dL), and low levels of high-density lipoprotein (HDL;<50 mg/dL in women,<40 mg/dL in men) consistent with the clinical criteria for metabolic syndrome were calculated for each racial/ethnic group. These five categorical variables were summed into a cardiometabolic burden variable (range 0–5) [25].

Neuroimaging

Brain scans are obtained on a 3T Siemens Magnetom SKYRA whole-body scanner and sequences of interest included a T1-weighted whole brain volumetric spoiled magnetization-prepared rapid gradient (1.1×1.1×1.2 mm; TR = 2300 ms; TE = 2.93 ms). All available MRI data was visually inspected for quality assurance and determined to have passed quality control (n = 1,294 for Meta-ROI and n = 1,194 WMH burden estimated) for inclusion in the present study.

Neurodegeneration of AD-vulnerable meta-ROI

FreeSurfer (version 5.3) was used to estimate intracranial volume, parcellate, and segment the T1-weighted scans. An AD-vulnerable meta-ROI composed of the surface-area weighted average of the mean cortical thickness in the entorhinal cortex, inferior temporal cortex, middle temporal cortex, middle temporal gyrus, and fusiform gyrus was to characterize neurodegeneration [26]. Lower AD vulnerable meta-ROI values are indicative of greater neurodegeneration or poorer brain health.

White matter hyperintensity (WMH) burden

SPM’s Lesion Segmentation Toolbox (www.applied-statistics.de/lst.html, SPM12) characterized white matter lesion volume. WMHs were identified on T2-weighted FLAIR images using the default settings of the lesion prediction algorithm [27]. Higher WMH burden values are indicative of greater white matter lesion pathology or poorer brain health.

Statistics

Analyses were performed in R version 3.5.0 (https://cran.r-project.org/). Serum FABP3 and t-tau were log transformed and WMH volume was Box-Cox transformed to improve normality. A one-way ANOVA and chi-squared analyses examined group differences in sample characteristics. Multiple linear regressions were used to 1) examine racial/ethnic group differences in plasma markers (FABP3, Aβ₄₂/Aβ₄₀ ratio, and t-tau) and 2) identify, independent of amyloid and tau, whether there were any FABP3 racial/ethnic interactions on WMH burden and neurodegeneration of the AD vulnerable meta-ROI within each racial/ethnic group. Covariates included age, education, sex, APOE ɛ4 positivity, MCI status, cardiometabolic burden, and intracranial volume (WMH burden only). Unstandardized beta estimates for predictors of interest are reported in the text. Multicollinearity statistics for all regression models were determined to be in the acceptable range (variance inflation factor <1.5, tolerance >0.85, all rs < 0.4).

RESULTS

Participant characteristics

Participant demographics and clinical characteristics are presented in Table 1. The Mexican American group was significantly younger, had fewer years of education, and had a greater proportion of women and individuals diagnosed with MCI relative to NHW group (ps < 0.001). The Mexican American group also had higher rates of cardiometabolic disease burden and performed more poorly on measures of general cognition and oral word reading (p < 0.001), but demonstrated lower rates of APOE ɛ4 positivity (ps < 0.001) relative to the NHW group.

Table 1

Participant demographics and clinical characteristics

	Non-Hispanic Whites (N = 676)	Mexican Americans (N = 693)	p
Age
Mean (SD)	69.1 (8.56)	63.3 (7.86)	<0.001
Median [Min, Max]	70.0 [50.0, 92.0]	63.0 [50.0, 90.0]
Education
Mean (SD)	15.5 (2.56)	9.81 (4.56)	<0.001
Median [Min, Max]	16.0 [0, 20.0]	10.0 [0, 20.0]
Sex
Males	297 (43.9%)	227 (32.8%)	<0.001
Females	379 (56.1%)	466 (67.2%)
APOE ɛ4 Positivity
ɛ4–	486 (71.9%)	574 (82.8%)	<0.001
ɛ4+	190 (28.1%)	119 (17.2%)
MCI Diagnosis
Cognitively Normal	596 (88.2%)	556 (80.2%)	<0.001
MCI	80 (11.8%)	137 (19.8%)
Vascular Risk Burden
Mean (SD)	1.74 (1.24)	2.54 (1.34)	<0.001
Median [Min, Max]	2.00 [0, 5.00]	2.00 [0, 5.00]
Geriatric Depression Score
Mean (SD)	4.39 (4.93)	6.20 (6.04)	<0.001
Median [Min, Max]	3.00 [0, 29.0]	4.00 [0, 29.0]
Missing	0 (0%)	1 (0.1%)
Annual Income ($)
Mean (SD)	86900 (82200)	38700 (52000)	<0.001
Median [Min, Max]	70000 [0, 1000000]	25000 [0, 800000]
Missing	12 (1.8%)	17 (2.5%)
American National Adult Reading Test (z-score)
Mean (SD)	0.224 (0.913)	–0.534 (1.00)	<0.001
Median [Min, Max]	0.300 [–3.00, 2.30]	–0.500 [–3.30, 2.00]
Missing	0 (0%)	438 (63.2%)
Mini-Mental Status Exam
Mean (SD)	29.1 (1.10)	26.5 (2.99)	<0.001
Median [Min, Max]	29.0 [24.0, 30.0]	27.0 [14.0, 30.0]

Racial/ethnic differences in plasma and neuroimaging markers

The Mexican American group had significantly lower levels of FABP3 (B = –0.04, standard error = 0.01, 95% C.I. [–0.07, –0.02], t = –3.89, p < 0.001) relative to the NHW group. However, the Mexican American group had significantly higher levels of Aβ₄₂/Aβ₄₀ (B = 0.004, standard error = 0.001, 95% C.I. [0.002, 0.006], t = 4.78, p < 0.001, $η_{p}^{2}$ = 0.02) and t-tau (B = 0.04, standard error = 0.01, 95% C.I. [0.01, 0.06], t = 3.06, p = 0.002) relative to the NHW group. See Fig. 1.

Fig. 1

Racial/Ethnic Group Differences in FABP3 Levels.

Racial/ethnic differences in plasma marker predictors of WMH burden and neurodegeneration

There were no significant FABP X race/ethnic group interactions on neurodegeneration (B = –0.07, standard error = 0.05, 95% CI [–0.15, 0.01], t = –1.71, p = 0.08) or WMH burden (B = –0.02, standard error = 0.11, 95% CI [–0.23, 0.20], t = –0.13, p = 0.89). However, a series of race-stratified multiple linear regression analyses were used to explore associations between plasma markers and neuroimaging metrics (see [28] for a discussion of the need to assess beyond interaction terms). Within the Mexican American group, results revealed that FABP3 was significantly negatively associated with neurodegeneration in the AD-vulnerable meta-ROI (B = –0.08, standard error = 0.03, 95% CI [–0.14, 0.03], t = –3.01, p = 0.003). See Fig. 2. Additionally, results revealed a significant positive association between FABP3 (B = 0.18, standard error = 0.08, 95% CI [0.02, 0.34], t = 2.25, p = 0.03) and WMH burden.

Fig. 2

FABP3 associations with neurodegeneration and WMH burden in Mexican Americans.

Within the NHW group, there were no significant associations with FABP3 (B = 0.02, standard error = 0.04, 95% CI [–0.05, 0.09], t = 0.52, p = 0.60) with neurodegeneration. Finally, no significant associations between FABP3 (B = 0.12, standard error = 0.09, 95% CI [–0.06, 0.30], t = 1.25, p = 0.21) and WMH burden were observed. Full regression tables for each race-stratified model are available in the supplemental materials.

DISCUSSION

Our findings revealed there were racial/ethnic differences in plasma FABP3 levels and that beyond traditional AD plasma biomarkers, FABP3 was significantly associated with WMH burden and neurodegeneration in Mexican Americans, but not NHWs. Although FABP3 levels were generally lower in Mexican Americans compared to NHWs, our findings suggest that FABP3 is significantly associated with poorer neural health in this group, highlighting its potential utility as an early prognostic marker in Mexican Americans.

While studies have highlighted that FABP may be an important biomarker in older adults [20, 21], this is the first known study to investigate ethnoracial differences in serum FABP levels. Independent of age, education, sex, APOE ɛ4 positivity, MCI diagnosis, and cardiometabolic burden, Mexican Americans demonstrated lower FABP3 levels relative to NHWs. This could be the result of ethnoracial variations in FABP single nucleotide polymorphisms that have previously been noted in another sample [29]. Additional work is needed to disentangle the potential role of genetic factors in the observed findings, but results highlight the need to explore ethnoracial differences in the general patterns of disease biomarkers within well-characterized and representative cohorts.

Although the accumulation of amyloid plaques and tau tangles are thought to drive gray and white matter neurodegeneration in AD, our findings highlight that that FABP3 may also play a contributory role. This aligns well with animal studies that have revealed that FABP expression is highly expressed the dentate gyrus and white matter [30, 31], and therefore illustrates that AD-vulnerable neural regions may be particularly susceptible to FABP changes.

While results revealed there were no significant FABP3 ethnoracial interactions on neural markers, race-stratified analyses revealed FABP3 was uniquely associated with neurodegeneration and WMH burden in the Mexican American group. Although another study in predominantly NHW samples demonstrated that higher baseline levels of FABP in cognitively normal older adults are predictive of future entorhinal atrophy [15], it is worthwhile to note the Mexican American group had higher levels of cardiometabolic disease burden, which has been intimately linked to lipid homeostasis [1]. Although we controlled for cardiometabolic burden, this variable was significantly associated with neural markers in our models. In other words, although FABP3 levels were generally lower levels within the Mexican American group, FABP3 could be more detrimental within the context of these negative cardiometabolic conditions given these also carry their own independent risk for poor neural health [32].

Importantly, independent of traditional AD markers, FABP3 was associated with neurodegeneration and WMH burden in the Mexican American group. This pattern of results raises questions about the mechanisms by which FABP3-related lipid homeostasis may cause brain changes. Although it is has been posited that FABP3 may promote neurodegeneration through impaired amyloid processing and clearance, amyloid was not linked to our neural metrics in this group [6]. Additional research is needed to clarify FABP3-associated neurodegenerative pathways, as inflammation and/or blood-brain barrier dysfunction may be alternative relevant pathways [6, 33].

Notable limitations of the study are that we do not have access to FABP genetic variants, and we did not have the ability explore if similar patterns emerge across other central nervous system FABPs (i.e., –5, –7). Our sample also consisted entirely of Mexican Americans, and it is important to note that the larger H/L population is highly heterogenous, and it is possible that these findings may not translate to other subethnic groups. Finally, while we focused one neurodegeneration in one AD-vulnerable meta-ROI, it is possible relationships with FABP may emerge in other brain regions. Strengths include utilizing the well-characterized HABS-HD cohort, with an emphasis on understanding the significance of these biomarkers within cognitively intact participants or prodromal phase of AD.

Conclusions

Plasma FABP3 may be important for understanding biological mechanisms underlying disparities in AD risk and development in racially/ethnically diverse older adults. Future studies should explore pathophysiological pathway by which FABP3 may initiate neurodegeneration and examine associations with functional neuroimaging metrics.

Footnotes

ACKNOWLEDGMENTS

The authors thank all the HABS-HD participants and study staff for their commitment to advancing representative aging research and for publicly sharing this data with other researchers.

Research reported in this publication was supported by the National Institute on Aging (NIA) of the National Institutes of Health under Award Numbers R01AG054073 and R01AG058533. The consent is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Drs. Clark, Duarte, and Haley received start-up from the University of Texas at Austin that supported this work. Dr. Clark is further supported by a Shiley Marcos Alzheimer Disease Research Education Center Grant (P30AG062429) from the NIA.

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Chew

, Solomon

, Fonteh

(2020) Involvement of lipids in Alzheimer’s disease pathology and potential therapies. Front Physiol 11, 598.

Liu

, Kanekiyo

, Xu

, Bu

(2013) Apolipoprotein E and Alzheimer disease: Risk, mechanisms and therapy. Nat Rev Neurol 9, 106–118.

Andrews

, Fulton-Howard

, Goate

(2020) Interpretation of risk loci from genome-wide association studies of Alzheimer’s disease. Lancet Neurol 19, 326–335.

O’Bryant

, Johnson

, Balldin

, Edwards

, Barber

, Williams

, Devous

, Cushings

, Knebl

, Hall

(2013) Characterization of Mexican Americans with mild cognitive impairment and Alzheimer’s disease. J Alzheimers Dis 33, 373–379.

González

, Tarraf

, Schneiderman

, Fornage

, Vásquez

, Zeng

, Youngblood

, Gallo

, Daviglus

, Lipton

, Kaplan

, Ramos

, Lamar

, Thomas

, Chai

, DeCarli

(2019) Prevalence and correlates of mildcognitive impairment among diverse Hispanics/Latinos: Study of Latinos-Investigation of Neurocognitive Agingresults. Alzheimers Dement 15, 1507–1515.

Sepe

, Chiasserini

, Parnetti

(2018) Role of FABP3 as biomarker in Alzheimer’s disease and synucleinopathies. Future Neurol 13, 199–207.

Bjerke

, Kern

, Blennow

, Zetterberg

, Waern

, Börjesson-Hanson

, Östling

, Kern

, Skoog

(2016) Cerebrospinal fluid fatty acid-binding protein 3 is related to dementia development in a population-based sample of older adult women followed for 8 years. J Alzheimers Dis |bf 49, 733–741.

Khan

, Aguilar

, Kiddle

, Doyle

, Thambisetty

, Muehlboeck

, Sattlecker

, Newhouse

, Lovestone

, Dobson

(2015) A subset of cerebrospinal fluid proteins from a multi-analyte panel associated with brain atrophy, disease classification and prediction in Alzheimer’s disease. PloS One 10, e0134368.

Steinacker

, Mollenhauer

, Bibl

, Cepek

, Esselmann

, Brechlin

, Lewczuk

, Poser

, Kretzschmar

, Wiltfang

(2004) Heart fatty acid binding protein as a potential diagnostic marker for neurodegenerative diseases. Neurosci Lett 370, 36–39.

10.

Oddo

, Caccamo

, Shepherd

, Murphy

, Golde

, Kayed

, Metherate

, Mattson

, Akbari

, LaFerla

(2003) Triple-transgenic model of Alzheimer’s disease with plaques and tangles: Intracellular Aβ and synaptic dysfunction. Neuron 39, 409–421.

11.

Pélerin

, Jouin

, Lallemand

M-S

, Alessandri

J-M

, Cunnane

, Langelier

, Guesnet

(2014) Gene expressionof fatty acid transport and binding proteins in the blood–brain barrier and the cerebral cortex of therat: Differences across development and with different DHA brain status. Prostaglandins Leukot Essent FattyAcids 91, 213–220.

12.

, Chen-Plotkin

, Arnold

, Grossman

, Clark

, Shaw

, Pickering

, Kuhn

, Chen

, McCluskey

(2010) Novel CSF biomarkers for Alzheimer’s disease and mild cognitive impairment. Acta Neuropathol 119, 669–678.

13.

Chiasserini

, Parnetti

, Andreasson

, Zetterberg

, Giannandrea

, Calabresi

, Blennow

(2010) CSF levels of heart fatty acid binding protein are altered during early phases of Alzheimer’s disease. J Alzheimers Dis 22, 1281–1288.

14.

Dulewicz

, Kulczyńska-Przybik

, Słowik

, Borawska

, Mroczko

(2021) Fatty acid binding protein 3(FABP3) and Apolipoprotein E4 (ApoE4) as lipid metabolism-related biomarkers of Alzheimer’s disease. J ClinMed 10, 3009.

15.

Desikan

, Thompson

, Holland

, Hess

, Brewer

, Zetterberg

, Blennow

, Andreassen

, McEvoy

, Hyman

, Dale

, Alzheimer’s Disease Neuroimaging Initiative (2013) Heart fatty acid binding protein and Abeta-associated Alzheimer’s neurodegeneration. Mol Neurodegener 8, 39.

16.

Babulal

, Quiroz

, Albensi

, Arenaza-Urquijo

, Astell

, Babiloni

, Bahar-Fuchs

, Bell

, Bowman

, Brickman

(2019) Perspectives on ethnic and racial disparities in Alzheimer’s disease and related dementias: Update and areas of immediate need. Alzheimers Dement 15, 292–312.

17.

Brickman

, Schupf

, Manly

, Luchsinger

, Andrews

, Tang

, Reitz

, Small

, Mayeux

, DeCarli

, Brown

(2008) Brain morphology in older African Americans, Caribbean Hispanics, and whites from northern Manhattan. Arch Neurol 65, 1053–1061.

18.

Matthews

, Xu

, Gaglioti

, Holt

, Croft

, Mack

, McGuire

(2019) Racial and ethnic estimates of Alzheimer’s disease and related dementias in the United States (2015-2060) in adults aged > /=65 years. Alzheimers Dement 15, 17–24.

19.

Filshtein

, Dugger

, Jin

, Olichney

, Farias

, Carvajal-Carmona

, Lott

, Mungas

, Reed

, Beckett

, DeCarli

(2019) Neuropathological diagnoses of demented Hispanic, Black, and Non-Hispanic White decedents seen at an Alzheimer’s disease center. J Alzheimers Dis 68, 145–158.

20.

O’Bryant

, Xiao

, Edwards

, Devous

, Gupta

, Martins

, Zhang

, Barber

(2013) Biomarkers of Alzheimer’s disease among Mexican Americans. J Alzheimers Dis 34, 841–849.

21.

O’Bryant

, Zhang

, Petersen

, Hall

, Johnson

, Yaffe

, Mason

, Braskie

, Barber

, Rissman

(2022) A blood screening tool for detecting mild cognitive impairment and Alzheimer’s disease among community-dwelling Mexican Americans and non-Hispanic Whites: A method for increasing representation of diverse populations in clinical research. Alzheimers Dement 18, 77–87.

22.

Institute for Translational Research, https://apps.unthsc.edu/itr/researchers.

23.

O’Bryant

, Johnson

, Barber

, Braskie

, Christian

, Hall

, Hazra

, King

, Kothapalli

, Large

(2021) The Health & Aging Brain among Latino Elders (HABLE) study methods and participant characteristics. Alzheimers Dement (Amst) 13, e12202.

24.

O’Bryant

, Gupta

, Henriksen

, Edwards

, Jeromin

, Lista

, Bazenet

, Soares

, Lovestone

, Hampel

, Montine

, Blennow

, Foroud

, Carrillo

, Graff-Radford

, Laske

, Breteler

, Shaw

, Trojanowski

, Schupf

, Rissman

, Fagan

, Oberoi

, Umek

, Weiner

, Grammas

, Posner

, Martins

, Star

, groups

(2015) Guidelines for the standardization of preanalytic variables for blood-based biomarker studies in Alzheimer’s disease research. Alzheimers Dement 11, 549–560.

25.

Eckel

, Grundy

, Zimmet

(2005) The metabolic syndrome. Lancet 365, 1415–1428.

26.

Jack

, Wiste

, Therneau

, Weigand

, Knopman

, Mielke

, Lowe

, Vemuri

, Machulda

, Schwarz

(2019) Associations of amyloid, tau, and neurodegeneration biomarker profiles with rates of memory decline among individuals without dementia. JAMA 321, 2316–2325.

27.

Schmidt

, Gaser

, Arsic

, Buck

, Förschler

, Berthele

, Hoshi

, Ilg

, Schmid

, Zimmer

(2012) An automated tool for detection of FLAIR-hyperintense white-matter lesions in multiple sclerosis. Neuroimage 59, 3774–3783.

28.

Ward

, Gartner

, Keyes

, Fliss

, McClure

, Robinson

(2019) How do we assess a racial disparity in health? Distribution, interaction, and interpretation in epidemiological studies. Ann Epidemiol 29, 1–7.

29.

Kim

, Kim

S-Y

, Kim

, Kwon

, rok Kim

, Lee

, Yoo

K-D

(2014) Polymorphisms of the heart-type fatty acid-binding protein as a prognostic factor in patients with atherosclerosis. Ann Clin Lab Sci 44, 67–72.

30.

Owada

(2008)Fatty acid binding protein: Localization and functional significance in the brain. Tohoku J Exp Med 214, 213–220.

31.

Murphy

, Owada

, Kitanaka

, Kondo

, Glatz

(2005) Brain arachidonic acid incorporation is decreased in heart fatty acid binding protein gene-ablated mice. Biochemistry 44, 6350–6360.

32.

Foret

, Dekhtyar

, Birdsill

, Tanaka

, Haley

(2021) Metabolic syndrome components moderate the association between executive function and functional connectivity in the default mode network. Brain Imaging Behav 15, 2139–2148.

33.

Brosseron

, Kleemann

, Kolbe

, Santarelli

, Castro-Gomez

, Tacik

, Latz

, Jessen

, Heneka

(2021) Interrelations of Alzheimer’s disease candidate biomarkers neurogranin, fatty acid-binding protein 3 and ferritin to neurodegeneration and neuroinflammation. J Neurochem 157, 2210–2224.