Epigenetics of dementia remains unraveled in Latin American and Caribbean populations: A call for collaborative efforts

Abstract

Dementia is a major health issue in Latin America and the Caribbean (LAC), with a prevalence of 9.5% and an incidence of 26.0 per 1000 among people over 60 years. With cases expected to triple by 2050, there is an urgent need for more extensive local research in this field.¹ Despite advancements in neuroimaging and protein biomarkers, significant gaps remain in understanding how biological mechanisms that interact with LAC-specific environmental exposures influence dementia risk, presentation, and treatment. Interactions between genetic and environmental factors, mainly through epigenetic changes including DNA methylation (DNAm), noncoding RNA, and histone modifications, can modulate gene expression, altering molecular traits and health outcomes.² Adverse environmental exposures in LAC, including socioeconomic disparities, pollutants, unhealthy habits, and comorbidities, have been associated with higher dementia risk,¹ potentially through epigenetic mechanisms. Most existing knowledge on epigenetics is derived from studies conducted in Europe and the United States, which limits the generalization of these findings to underrepresented regions,³ including LAC. To understand state-of-the-art epigenetic studies on dementia in LAC, we conducted a systematic review following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) protocols. Our search focused on chemical modifications of the DNA or histones able to regulate chromatin's structure. We reviewed the literature up to May 2024 in MEDLINE, Web of Science, PubMed, and Scopus databases (research strategy and findings in Table 1). Surprisingly, we found only five case-control studies comparing late-onset Alzheimer's disease (LOAD)^4-6 or mild cognitive impairment (MCI)^{7, 8} versus healthy controls, revealing a disparity in published studies on epigenetics and dementia in LAC. Further epigenetics research is needed in the region, including the use of cutting-edge methods, target tissues, and systematic approaches. Studies have investigated global DNAm (LINE-1).⁴ DNAm in candidate genes.^{5, 6} or genome-wide DNAm^{7, 8} assessed in peripheral blood. Two studies were conducted in Colombian,^{4, 5} two in Mexican-American (MA),^{7, 8} and one in Costa Rican populations.⁶ Three studies addressed sex^{4, 5} or ethnic⁸ differences in DNAm. The study conducted by Hernández et al. found no significant differences in LINE-1 methylation across LOAD patients and controls, even after stratification by sex or APOE4 genotypes.⁴ Further, Salcedo-Tacuma et al. identified significantly lower DNAm levels at three CpGs at BIN1 gene in LOAD patients.⁵ In 2019, Pathak et al. reported differentially methylated CpGs in MCI individuals in an Epigenome-Wide Association Study (EWAS). These findings were linked to neuronal death, metabolic dysfunction, and inflammation.⁷ In 2021, Coto-Vílchez et al. compared Horvath's epigenetic clock DNAm profiles to measure biological aging. Both LOAD patients and control group exhibited an average epigenetic age 20 years younger than their chronological age. Additionally, they identified differentially methylated regions at PM20D1 across study groups.⁶ Finally, Abraham Daniel et al. analyzed ethnicity-specific DNAm profiles in non-Hispanic whites and MA, identifying significant differential methylation at CREBBP gene associated with MCI and AD in MA participants.⁸ These findings generally agree with previous evidence from other populations, suggesting distinctive DNAm patterns associated with dementia or impaired cognition, although some contradictions remain regarding differential epigenetic aging. Also, they provide evidence showing that genetic ancestry influences DNAm patterns. Overall, limitations of these studies include restricted sample size,^4-7 limited follow-up data for functional interpretation like gene expression, and the use of peripheral blood-derived methylation biomarkers while they may not reflect brain-specific alterations in dementia. Unfortunately, none of the reviewed articles investigated histone modifications.

TABLE 1. Characteristics of the selected studies

Authors and year	LAC Population	Sample size	Epigenetic measurement examined	Outcome measure
Hernández et al., 2013	Colombia	28 LOAD and 30 controls. Stratification by sex and ApoE4 allele.	Global DNAm methylation (MS-HRM)	No significant differential methylation at LINE-1 gene.
Salcedo-Tacuma et al., 2019	Colombia	50 LOAD and 50 controls. Stratification by age, sex, and ApoE4 allele.	Candidate gene (Bisulfite conversion and PCR)	LOAD patients showed significantly lower DNAm at three CpGs at BIN1 gene: CpG26, CpG44, and CpG87.
Pathak et al., 2019	Mexican American	45 MCI and 45 controls. MA has a high metabolic risk score, less years of education and a lower frequency of the ApoE4 allele.	EWAS (Infinium MethylationEPIC BeadChip)	Four CpG sites significantly hypomethylated in MCI subjects compared to controls: cg25016219 (KLHL29 gene), cg26479998 (SEPT9 gene), cg02586267 (not mapped), and cg18978297 (CPLX3 gene). Six CpG sites significantly hypermethylated in MCI subject compared to controls: cg22360048 (PKIB gene), cg20904111 (intergenic), cg05917713 (BCL2L2-PABPN1 gene), cg20201669 (OR2C3 gene), cg14179796 (CCNY gene), cg22327037 (intergenic).
Coto-Vílchez et al., 2021	Costa Rica	11 LOAD and 21 controls. All female participants were over 90 years old, with educational backgrounds ranging from 0 to 9 years.	EWAS (Infinium MethylationEPIC BeadChip) and Horvath´s epigenetic clock	The LOAD patients and the control group had an epigenetic age of 20 years younger than their chronological age. Significantly higher methylation at the PM20D1 gene in individuals who were heterozygotes (AG) for rs708727 than GG genotype individuals.
Abraham Daniel et al. 2023	Mexican American	122 AD/MCI and 177 controls. Stratification by sex, age, years of education, and ApoE4 allele	EWAS (Infinium MethylationEPIC BeadChip)	Significantly hypermethylated CpG site at CREBBP (cg13135255) in cognitively impaired participants (AD/MCI) compared to controls.

• Note: For dementia, the search terms used were: “alzheimer's disease,” “AD,” “dementia,” “frontotemporal dementia,” “FTD,” “frontotemporal lobar degeneration,” “apoliprotein e,” and “ApoE.” For epigenetics, the terms used included: “epigenetic,” “epigenomic,” “DNA methylation,” “histone modification,” “epigenome-wide association study,” and “EWAS”. The country-specific terms included the names of all countries within Latin America and the Caribbean (LAC).
• Abbreviations: DNAm, DNA methylation; EWAS, Epigenome-Wide Association Study; LOAD, late-onset Alzheimer's disease; MCI, mild cognitive impairment; MS-HRM, methylation-sensitive high-resolution melting.

Despite the limited number of identified studies, their positive quality assessment by the Newcastle-Ottawa scale (6 to 8 points) suggests considerable potential for conducting meaningful research and providing valuable insights into the epigenetic landscape of dementia within the region. Given the scarce but encouraging evidence, it is imperative to promote further efforts to unravel LAC-specific epigenetic biomarkers of dementia. We advocate for coordinated and systematic joint efforts to guide future studies relevant to the context of the region.

In conclusion, we call for actions to address the gaps in epigenetics-dementia research in LAC by:

1. Stimulating local epigenetics research and leveraging consortia for the study of dementia, like ReDLat, LAC-CD, or UNITED Consortium,^{9, 10} to enhance collaboration, interdisciplinarity, and harmonization across studies.

2. Assessing the research landscape of the region, to identify gaps, strengths, limitations, and opportunities. This knowledge will serve for the planification of further efforts in the field.

3. Strengthening local research capacities by empowering LAC researchers through tailored training and funding opportunities.

We believe these efforts are essential for understanding the role of gene-environment interplay in dementia development in LAC.