Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jan 1.
Published in final edited form as: Nurs Res. 2024 Oct 10;74(1):37–46. doi: 10.1097/NNR.0000000000000783

Metabolic Pathways Associated with Obesity and Hypertension in Black Caregivers of Persons Living with Dementia

Glenna S Brewster 1,*, Madelyn C Houser 2,*, Irene Yang 3, Jordan Pelkmans 4, Melinda Higgins 5, Cristy Tower-Gilchrist 6, Jessica Wells 7, Arshed A Quyyumi 8, Dean Jones 9, Sandra Dunbar 10, Nicole Carlson 11
PMCID: PMC11637965  NIHMSID: NIHMS2023197  PMID: 39420455

Abstract

Background:

In the U.S., Black adults have the highest prevalence of obesity and hypertension, increasing their risk of morbidity and mortality. Caregivers of persons with dementia are also at increased risk of morbidity and mortality due to the demands of providing care. Thus, Black caregivers—who are the second largest group of caregivers of persons with dementia in the U.S.—have the highest risks for poor health outcomes among all caregivers. However, the physiologic changes associated with multiple chronic conditions in Black caregivers are poorly understood.

Objectives:

In this study, metabolomics were compared to the metabolic profiles of Black caregivers with obesity, with or without hypertension. Our goal was to identify metabolites and metabolic pathways that could be targeted to reduce obesity and hypertension rates in this group.

Methods:

High-resolution, untargeted metabolomic assays were performed on plasma samples from 26 self-identified Black caregivers with obesity, 18 of whom had hypertension. Logistic regression and pathway analyses were employed to identify metabolites and metabolic pathways differentiating caregivers with obesity only and caregivers with both obesity and hypertension.

Results:

Key metabolic pathways discriminating caregivers with obesity only and caregivers with obesity and hypertension were butanoate and glutamate metabolism, fatty acid activation/biosynthesis, and the carnitine shuttle pathway. Metabolites related to glutamate metabolism in the butanoate metabolism pathway were more abundant in caregivers with hypertension, while metabolites identified as butyric acid/butanoate and R-(3)-hydroxybutanoate were less abundant. Caregivers with hypertension also had lower levels of several unsaturated fatty acids.

Discussion:

In Black caregivers with obesity, multiple metabolic features and pathways differentiated among caregivers with and without hypertension. If confirmed in future studies, these findings would support ongoing clinical monitoring and culturally tailored interventions focused on nutrition (particularly polyunsaturated fats and animal protein), exercise, and stress management to reduce the risk of hypertension in Black caregivers with obesity.

Keywords: African American, Alzheimer’s disease, diet, hypertension, metabolomics

Black adults are the second largest group of caregivers of persons living with dementia (PLwD), comprising approximately 10% of the total caregiver population (Alzheimer’s Association, 2024). Although the service they provide benefits both PLwD and society, it often leaves caregivers with little time for self-care (García-Martín et al., 2023). Poor self-care can result in weight gain or obesity, which is more prevalent among caregivers than noncaregivers (Horner-Johnson et al., 2015). Caregivers also have a high prevalence of multiple chronic conditions (MCC), with nearly 60% diagnosed with more than two chronic conditions, such as hypertension, arthritis, or heart disease (Wang et al., 2014). Irrespective of caregiver status, Black populations experience a higher prevalence and severity of hypertension compared to White adults (Carson et al., 2011). Therefore, Black caregivers of PLwD are more likely to experience both obesity and hypertension, which could be attributed to genetic and environmental factors such as diet, lifestyle, and caregiving responsibilities (Baik et al., 2023; Knight et al., 2007).

While people who are overweight or obese are more likely to be diagnosed with hypertension (Landi et al., 2018), a proportion of individuals with obesity do not develop MCC and maintain a relatively healthy metabolic state (Hinnouho et al., 2015). These metabolically healthy people with obesity have lower mortality rates (Hinnouho et al., 2015). Progression from obesity only to obesity and hypertension is difficult to predict, which limits the effectiveness of clinical interventions in this population. To reduce the risk for development of MCC, researchers and clinicians require more information on the physiologic mechanisms of obesity and hypertension in Black caregivers.

Untargeted metabolomics is used to assess relative levels of a large percentage of the small molecules, or metabolites, in biological samples. Metabolites are generated via the activity of a series of linked biochemical reactions, or pathways, responsible for the execution of basic biological processes such as the production of amino acids. The concentrations of metabolites in cells, tissues, and various bodily fluids are influenced by a person’s genetic predisposition, environmental exposures, and interactions between the two as well as by the activity of resident microbes (Tahir et al., 2022). As such, metabolomics is a useful methodology for studying the pathophysiology underlying MCC.

Researchers have identified several metabolic pathways that distinguish metabolically healthy individuals with obesity from those who are not. These pathways include fatty acid biosynthesis, pyrimidine metabolism, galactose metabolism, valine, leucine and isoleucine degradation, tryptophan metabolism, propanoate metabolism, phenylalanine metabolism, glyoxylate and dicarboxylate metabolism, and the citrate cycle (Chen et al., 2015). These differences in metabolic pathways better predict metabolic health than the degree of obesity. Many of these processes are crucial for energy balance, mitochondrial function, and managing oxidative stress, all of which influence adipocyte differentiation and function (de Mello et al., 2018). Researchers have also made strides in understanding metabolic changes associated with hypertension, including alterations in inflammation and oxidative stress-related pathways, as well as shifts in the metabolism of fatty acids, steroid hormones, amino acids, and glucose (Onuh & Qiu, 2021). These pathways suggest that energy metabolism or hormone level disruptions may trigger inflammation and oxidative stress, contributing to insulin resistance, endothelial dysfunction, and vascular remodeling—key factors in hypertension development. Few studies have directly linked these metabolic changes to specific causes. Potential contributors could include variations in genetically or epigenetically regulated enzyme levels or activity, microbiome composition and function, physical activity levels, and dietary factors such as protein and dairy intake (Chen et al., 2015; Onuh & Qiu, 2021).

While a robust body of research addressing the relationship between metabolism and MCC, such as obesity and hypertension, is emerging, many metabolites and metabolic pathways differentiating people with these conditions have yet to be identified. In addition, the mechanisms by which these metabolites contribute to obesity/hypertension are still largely unclear. There are limited studies that have focused on the highest risk group: Black adults, including caregivers of PLwD, and very few that have examined hypertension-related metabolites in the context of obesity. This study was designed as a first step to address this gap in knowledge, comparing the metabolic profiles of Black caregivers with obesity to the profiles of similar caregivers with obesity and hypertension. We used high-throughput, untargeted metabolomics to identify metabolites and metabolic pathways that should be prioritized for further investigation as potential targets for interventions to reduce MCC and enhance quality of life in this population.

Methods

Study Design and Ethical Approval

This was a prospective study with Black family caregivers of PLwD living in the state of Georgia. The Emory University Institutional Review Board approved this study.

Participants

To be included in the study, caregivers had to be (a) caring for PLwD; (b) self-identify as Black or African American; (c) between 30 and 85 years of age; and (d) meet the criteria for obesity (BMI ≥ 30 kg/m2). To be classified as having hypertension, participants had to report a diagnosis of hypertension or be currently taking medication for hypertension. Caregivers were excluded if (a) they had an uncontrolled major mental disorder (i.e., schizophrenia, bipolar disorder, major depression); (b) they had uncontrolled severe hypertension (systolic ≥ 200 mmHg or diastolic ≥ 110 mmHg) due to the emergent nature of this diagnosis; or (c) PLwD were living in an assisted living facility, nursing home, or another institutional setting.

Anthropometrics

Trained research staff collected and calculated body mass index (BMI) as body weight (kg)/body height (m)2.

Comorbidity

The Charlson Comorbidity Index is a weighted index morbidity score that assesses the number and severity of 19 conditions that may influence mortality risk (Charlson et al., 1987).

Sample Collection

Participants underwent venipuncture and collection of 5 mL whole blood into EDTA-containing tubes (Vacutainer, Becton Dickinson) by a nurse or phlebotomist. Tubes were centrifuged at 2,000 x g for 10 min at 4°C. Aliquots of plasma supernatant in clean polypropylene tubes were stored at −80°C until processed for metabolomics analysis.

High-Resolution Untargeted Metabolomics

After thawing, plasma samples were analyzed by the Emory Clinical Biomarkers Laboratory using a high-resolution, untargeted metabolomics platform (Liu et al., 2020). Briefly, plasma samples were treated with acetonitrile (2:1, v/v), spiked with internal standard mix, depleted of larger proteins by centrifugation at 14,000 × g for 5 min at 4°C, and assayed with liquid chromatography-mass spectrometry on a ThermoFisher Orbitrap Fusion Tribrid mass spectrometer (https://www.thermofisher.com/us/en/home.html). Samples were randomized with respect to hypertension status to minimize batch effects and run in triplicate with reference standards at the beginning, middle, and end of each run. To ensure capture of a broader range of metabolites, samples were assayed by both hydrophilic interaction liquid chromatography (HILIC) with positive electrospray ionization (ESI; HILIC+) and C18 hydrophobic reversed-phase chromatography with negative ESI (C18-). The two resulting data sets were analyzed separately.

Noise removal and feature extraction, alignment, quantification, and quality evaluation were performed as previously described (Liu et al., 2020). A set of confirmed metabolites and internal standards were used to ensure data quality with respect to number of features detected, missing values, and mass accuracy (threshold: < 5 ppm). Features were distinguished by mass-to-charge ratio (m/z) and retention time (RT). Triplicate technical replicates were median summarized.

Power Analysis

We used the R package pwr (Champely, 2017) to evaluate the statistical power for various effect sizes based on Cohen’s guidelines (1988), which classify effect sizes of 0.02, 0.15, and 0.35 as small, medium, and large, respectively. These effect sizes were derived from the partial correlation for the parameter being tested. For an alpha of 0.05 and a sample size of 30, the power to detect small, medium, and large effect sizes was 0.11, 0.49, and 0.84, respectively.

Data Analysis

Wilcoxon rank-sum tests (continuous variables) and Fisher’s exact test (categorical variables) were used to compare demographics and health characteristics in participants with and without hypertension. xmsPANDA (v1.3.2; https://github.com/kuppal2/xmsPANDA) was used to analyze metabolomics data. Data were filtered to retain only features found in at least 80% of samples from participants with or without hypertension. Feature intensities below detection level were replaced with half of the minimum value in the entire data set. Log2 transformation was applied to reduce heteroscedasticity and quantile normalization to reduce between-sample variability. Features that differed in abundance between participants with and without hypertension were selected by logistic regressions (p < .05), where hypertension status was the binary dependent variable, and transformed, normalized intensity of a feature was the predictor, and [|log2(fold change)| ≥ 0.5]. Due to the exploratory nature of this study and the goal of identifying candidate metabolites for further investigation in larger cohorts, no correction for multiple comparisons was implemented.

To identify metabolic pathways represented by the differentially abundant features, we utilized the mummichog Python tool (v1.0.10) with 1,000 permutations, primary ion required, and p < .05 as the criterion for features differing between participants with versus without hypertension (Li et al., 2013). Only pathways with adjusted p < .05 (adjustment accounted for degree of metabolite differential abundance and number of differentially abundant features relative to total features assigned to the pathway) and containing at least four differentially abundant metabolites were retained. The KEGG PATHWAY database (https://www.genome.jp/kegg/pathway.html) was consulted for information about relationships among metabolites in the selected pathways.

A reference library of more than 200 common metabolites (Liu et al., 2020) produced by the Emory Clinical Biomarkers Laboratory was used for feature annotation (m/z within 5, RT within 30s), as were annotations from mummichog and those derived at medium or high confidence from the R package xMSannotator (v1.3.2; Uppal et al., 2017). Annotations concordant between at least two of the three annotation methods were considered most credible.

Results

Participants

Twenty-six Black caregivers of PLwD participated in the study and provided blood samples (Table 1); among these, 18 had hypertension. Mean age of participants was 62.9 years, and two thirds were retired. The vast majority (88.5%) were female, half were married, and 63% had a bachelor’s degree or higher. More than half were providing care for their mothers, and most (73.8%) lived in the same house as the PLwD. Participants had been providing care for an average of 4.33 years. None of the demographic or health characteristics evaluated differed between participants with or without hypertension except that participants with hypertension reported more comorbidities (0–6, mean = 1.4) than those without (0–1, mean =.13, p = .01).

Table 1.

Participant Demographic and Health Characteristics

Hypertension (n=18) No Hypertension (n=8) Total (N=26) p value
Age (in years): .845 1
 Mean (SD) 62.833 (6.195) 63.000 (7.445) 62.885 (6.452)
 Minimum-Maxiumn 47.000 – 70.000 51.000 – 72.000 47.000 – 72.000
Sex 1.000 2
 Female 16 (88.9%) 7 (87.5%) 23 (88.5%)
 Male 2 (11.1%) 1 (12.5%) 3 (11.5%)
Marital Status .662 2
 Divorced or separated 7 (38.9%) 3 (37.5%) 10 (38.5%)
 Married or partnered 7 (38.9%) 5 (62.5%) 12 (46.2%)
 Never married 3 (16.7%) 0 (0.0%) 3 (11.5%)
 Widowed 1 (5.6%) 0 (0.0%) 1 (3.8%)
Education .360 2
 Associate’s degree or higher 11 (61.1%) 7 (87.5%) 18 (69.2%)
 Less than associate’s degree 7 (38.9%) 1 (12.5%) 8 (30.8%)
Employment .822 2
 Retired or Disabled 13 (72.2%) 5 (62.5%) 18 (69.2%)
 Unemployed 1 (5.6%) 1 (12.5%) 2 (7.7%)
 Working 4 (22.2%) 2 (25.0%) 6 (23.1%)
Total number of persons in household: .977 1
 Mean (SD) 2.556 (1.042) 2.500 (0.926) 2.538 (0.989)
 Range 1.000 – 5.000 1.000 – 4.000 1.000 – 5.000
Charlson Comorbidity Number (count of comorbidities) .010 1
 Mean (SD) 1.389 (1.501) 0.125 (0.354) 1.000 (1.386)
 Minimum-Maximum 0.000 – 6.000 0.000 – 1.000 0.000 – 6.000
Diabetes that requires treatment .132 2
 No 12 (66.7%) 8 (100.0%) 20 (76.9%)
 Yes 6 (33.3%) 0 (0.0%) 6 (23.1%)
Moderate or severe renal disease 1.000 2
 No 16 (88.9%) 8 (100.0%) 24 (92.3%)
 Yes 2 (11.1%) 0 (0.0%) 2 (7.7%)
Open in a new tab

Note.

1

Wilcoxon rank-sum test;

2

Fisher’s exact test

C18- Metabolomics

From the C18- untargeted metabolomics assay, 12,746 features were detected, with 10,060 features remaining after filtering. Using logistic regression, we identified 474 features that differed in abundance (p < .05) between Black caregivers with obesity and hypertension versus those with only obesity. Among these, 88 were more abundant (log2(fold change) ≥ 0.5) in participants with obesity and hypertension, and 16 were less abundant (Figure 1A).

Figure 1.

Figure 1.

Open in a new tab

C18- metabolites differentially abundant in Black caregivers with obesity and hypertension compared to those without hypertension. Figure 1A) Volcano plot of C18- metabolomic features, with fold-change (log2) in normalized metabolomic feature intensity in participants with hypertension compared to normotensive participants vs. −log10(p value) derived from logistic regression. Points represent features with raw p < .05 and log2(fold change) ≥ 0.5 (orange) or log2(fold change) ≤ −0.5 (blue) in participants with hypertension. Figure 1B) Metabolic pathways represented by the C18- metabolites differentially abundant between participants with and without hypertension. Pathways identified by mummichog (v1.0.10) with adjusted p < .05 and containing at least four differentially abundant metabolites. Orange and blue bars indicate that at least 80% of the differentially abundant metabolites associated with that pathway were more or less abundant, respectively, in participants with hypertension vs those without. Gray bars indicate that a mixture of metabolites more and less abundant in participants with hypertension were associated with that pathway. Figure 1C) Pathway diagram representing branches of butanoate and glutamate metabolic pathways that are predicted to be differentially active in participants with vs without hypertension. Circles represent C18- metabolites more (orange) or less (blue) abundant in participants with hypertension. Pathway information from KEGG PATHWAY database. Figure 1D) C18- fatty acids less abundant (blue circles) in participants with hypertension. Purple circle indicates that two different putative metabolites were annotated as the same molecule, one that was more abundant in participants with hypertension, and one that was less abundant. HTN – hypertension.

These 474 C18- features were linked to 17 metabolic pathways enriched in differentially abundant metabolites and thus likely differed in activity between the participant groups (Figure 1B). We focused further attention on four of these 17, namely butanoate metabolism, glutamate metabolism, fatty acid activation, and de novo fatty acid biosynthesis because the majority of differentially abundant metabolomic features assigned to these pathways were annotated identically by two different annotation methods (Supplemental Digital Content [SDC], SDC 1 and SDC 2), increasing confidence in the identification of these putative metabolites. Many other identified pathways were related to these four (SDC 1).

Butanoate and Glutamate Metabolism

With the notable exceptions of butanoate and R-(3)-hydroxybutanoate, differentially abundant metabolites assigned to the butanoate metabolism pathway were found at higher levels in participants with obesity and hypertension compared to normotensive participants (Figures 1B–C, SDC 2). These higher-abundance metabolites were primarily glutamate and its derivatives, including succinate, and were linked to both butanoate and glutamate metabolism. Features annotated as pyruvate and S-(2)-acetolactate were also more abundant in participants with hypertension (Fig 1C, SDC 2).

Fatty Acid Activation and Biosynthesis

Most of the differentially abundant metabolites assigned to the fatty acid activation and biosynthesis pathways were less abundant in participants with hypertension (Figures 1B and 1D, SDC 2). These included features annotated as omega-3 fatty acids alpha-linolenic and docosahexaenoic acid, omega-6 fatty acids linoleic, gamma-linolenic, and dihomo-gamma-linolenic (DGLA) acids, and omega-9 oleic acid, as well as butyric acid, stearic acid, and arachidic acid. One feature tentatively annotated as DGLA was significantly less abundant in normotensive participants (beta: −1.24; CI: −2.65, −.17; p = .04; Figure 1D, SDC 2).

HILIC+ Metabolomics

HILIC+ data showed 21,262 features, with 10,831 remaining after filtering. Among these, 946 features differed between the two groups by logistic regression (p < .05). As in the C18- data, most of the differentially abundant features were more abundant in participants with obesity and hypertension (Figure 2A). Five metabolic pathways were represented by the differentially abundant HILIC+ features (Figure 2B, SDC 3). Each of these pathways included both metabolites that were more abundant in participants with hypertension and metabolites that were less abundant. Unlike in the C18- data, for no pathway were the annotations for most differentially abundant metabolites congruent between the two annotation methods used, although annotations were congruent for several metabolites assigned to the carnitine shuttle pathway (SDC 3 and 4). Furthermore, carnitine shuttle activity, linoleate metabolism, and urea cycle/amino group metabolism pathways were identified by analyzing both the C18- and HILIC+ data sets.

Figure 2.

Figure 2.

Open in a new tab

HILIC+ metabolites differentially abundant in Black caregivers with obesity and hypertension compared to those without hypertension. Figure 2A) Volcano plot of HILIC+ metabolomic features, with fold-change (log2) in normalized metabolomic feature intensity in participants with hypertension compared to normotensive participants vs. −log10(p value) derived from logistic regression. Points represent features with raw p<.05 and log2(fold change) ≥ 0.5 (orange) or log2(fold change) ≤ −0.5 (blue) in participants with hypertension. Figure 2B) Metabolic pathways represented by the HILIC+ metabolites differentially abundant between participants with and without hypertension. Pathways identified by mummichog (v1.0.10) with adjusted p<.05 and containing at least four differentially abundant metabolites. Gray bars indicate that a mixture of metabolites more and less abundant in participants with hypertension were associated with that pathway. HTN – hypertension.

Discussion

This study compared metabolic profiles of Black caregivers with obesity and with obesity and hypertension. Key pathways differentiating these groups were butanoate/glutamate metabolism and fatty acid activation/biosynthesis. These findings were considered the most robust, as two different annotation methods corroborated the annotations of metabolites in these pathways. Additionally, the carnitine shuttle pathway, linoleate metabolism, and urea cycle/amino group metabolism were identified as differentiators of hypertensive status from analyses of C18- and HILIC+ data sets, supporting the reliability of these findings.

Butanoate/Glutamate Metabolism

Several metabolomic features more abundant in participants with obesity and hypertension were annotated as glutamate and its direct derivatives or precursors, as well as other molecules involved in glutamate metabolism. The amino acid glutamate mediates numerous physiological functions on its own, and it can also serve as a substrate to produce other amino acids, nucleotides, and fatty acids, including butyric acid and its conjugate base butanoate (Yelamanchi et al., 2016). Glutamate can function as a neurotransmitter; its role in the central regulation of blood pressure has been well-established (Gabor & Leenen, 2012). Increased glutamate release in key regions of the brain results in increased blood pressure (Gabor & Leenen, 2012). This pattern extends beyond the central nervous system (CNS) as well, with elevated peripheral glutamate levels documented in clinical studies of hypertension (Dumas et al., 2018).

Diet is a primary determinant of glutamate levels, with common sources including dairy products, red meat, fish, and poultry (Ma et al., 2018). While often a minor dietary contributor, monosodium glutamate intake has also been associated with high blood pressure in human studies (Shi et al., 2011), and changes in glutamate metabolism have been linked to hypertension in the context of high salt diets (Shi et al., 2022). A randomized, controlled dietary intervention trial in Spain found that plasma levels of glutamate at baseline predicted risk of cardiovascular events and stroke the following year. Individuals with high baseline glutamate on a Mediterranean diet during the study had a reduced risk of cardiovascular events compared with those who did not receive the dietary intervention (Zheng et al., 2016).

In addition to diet, exercise can partially mitigate hypertension through modulation of glutamate signaling (Zhang et al., 2016). There is also some evidence that meditation may influence glutamate release in the brain (Fayed et al., 2013). Both exercise and meditation are often applied as measures to relieve stress. Recent studies have shown that chronic stress perturbs glutamate metabolism and signaling in the CNS (Kim & Jeon, 2018). However, much remains to be learned about the relationships among dietary glutamate intake, peripheral glutamate levels and metabolism, CNS glutamate levels and neurotransmission, and the physiological effects of different activities and exposures. This pilot study indicates that research in this area could potentially benefit Black caregivers, who frequently experience high levels of chronic stress and have limited time for self-care. These factors likely contribute to increased risks of MCC partly through dysregulation of glutamate metabolism.

Fatty Acid Activation/Biosynthesis and Linoleate Metabolism

In this study, Black caregivers with obesity and hypertension had lower plasma levels of metabolites annotated as fatty acids compared to similar caregivers without hypertension. These included polyunsaturated, monounsaturated, and saturated fatty acids. Polyunsaturated fatty acids less abundant in participants with hypertension were omega-3 and omega-6 fatty acids. Several of these kinds of molecules are components of the linoleate metabolism pathway and have known antihypertensive effects (Miura et al., 2008). Levels of omega-3 and omega-6 fatty acids in circulation reflect dietary intake. Nuts, seeds, vegetable and nut oils, and certain fish are typically the dietary sources of these molecules.

One monounsaturated fatty acid was found to be less abundant with hypertension: oleic acid. Omega-9 fatty acids like oleic acid can be produced in the body, but oleic acid is also the chief component of olive oil and the primary mediator of olive oil’s beneficial effects on blood pressure (Terés et al., 2008). Oleic acid was previously identified in metabolomics studies of hypertension, but in a study with 70–80-year-old Swedish adults (Lin et al., 2020), few of whom were obese, oleic acid was positively correlated with blood pressure. Given these differing results in distinct populations and the known vasodilatory activity of oleic acid, further study of the relationship between oleic acid and hypertension is needed.

A few saturated fatty acids were also less abundant in caregivers with obesity and hypertension. These included long-chain stearic acid and its elongation product: arachidic acid. Arachidic acid is found in corn oil, peanut oil, and cocoa butter. Dietary sources of stearic acid include animal fats and, to a lesser extent, vegetable fats, though it is abundant in coconut oil, shea butter, and cocoa butter. As these latter are common ingredients in moisturizing skin and hair care products, differences in stearic and arachidic acid levels in plasma samples in this study may derive from differential cosmetic exposure rather than differences in diet. While many saturated fatty acids have been linked to poor cardiovascular health, studies have found negligible-to-beneficial effects of dietary stearic acid consumption on blood pressure, total cholesterol, and triglyceride levels in circulation (Kris-Etherton et al., 2005). Fewer studies have examined the physiological effects of arachidic acid, but an inverse association with cardiometabolic morbidity and mortality has been found (Bockus et al., 2021). Thus, the relative abundance of these long-chain saturated fatty acids in normotensive participants is not incongruous with maintenance of cardiovascular health.

Another less abundant saturated fatty acid in participants with hypertension was butyric acid (with its conjugate base butanoate). Butyric acid is a short-chain fatty acid primarily produced by intestinal bacteria during carbohydrate fermentation. Other metabolomics studies have reported reduced plasma butyric acid in individuals with hypertension (Calderón-Pérez et al., 2020) and lower levels of butyric acid derivatives in individuals classified as metabolically healthy obese relative to those with obesity and metabolic syndrome (Wei et al., 2023). Butyric acid levels are inversely associated with blood pressure in human studies (Tilves et al., 2022). One mechanism that may drive this inverse association may be afferent vagal stimulation elicited by intestinal butyric acid that stimulates glutamate release in the CNS and its resultant hypotensive regulation (Cookson, 2021). Differential abundance of butyric acid with hypertension suggests that, in addition to potential diet differences, differences in the composition and activity of intestinal bacterial populations may contribute to cardiovascular health outcomes in Black caregivers with obesity, highlighting an avenue for future study. Finally, culturally tailored interventions related to diet (Kramer et al., 2023)—particularly to polyunsaturated fats and protein consumption—exercise (Bland & Sharma, 2017) and stress management may help optimize glutamate metabolism and fatty acid profiles in Black caregivers and thus reduce their risk of developing obesity and hypertension.

Carnitine Shuttle Metabolites and Urea Cycle/Amino Group Metabolism

Given the differences in fatty acid profiles by hypertension status, it is not surprising that the carnitine shuttle pathway—which is responsible for transporting long-chain fatty acids into the mitochondrial matrix in energy production—was identified as a pathway differentiating the two participant groups. Many of the differentially abundant fatty acids utilize the carnitine shuttle to cross the inner mitochondrial membrane. A few studies have linked circulating acylcarnitine levels with elevated blood pressure and hypertension (Arjmand et al., 2023). Additionally, as levels of carnitine and its derivatives are affected by the consumption of animal products, differences in dietary habits between the groups could influence the abundance of components of the carnitine shuttle. Similarly, urea cycle/amino group metabolism differences were identified in analysis of both C18- and HILIC+ data sets, and catabolism of amino acids is a significant source of energy during times of fasting or when amino acids are ingested at amounts higher than those required by the body to build tissues (Torres et al., 2023). These findings further suggest that hypertension among Black caregivers with obesity may be linked to diets high in protein.

Limitations

The small sample size limits our ability to implement multiple comparison adjustments or to control for potential covariates. Because caregivers in this study were only Black or African American adults, these results may not be generalizable to people of different racial/ethnic backgrounds. Dietary and medication intake, as well as time of day for sample collection, can affect metabolomic results; these variables were not accounted for in this analysis. Moreover, some participants were on antihypertensive medications that have the potential to mask or hide some of the metabolic features of hypertension. Finally, untargeted metabolomics relies on post hoc annotation to identify detected features; these annotations must be considered tentative until validated in targeted assays.

Conclusion

The findings of this study provide potential targets for interventions aimed at improving cardiovascular health in Black caregivers of PLwD, a group with disparate outcomes. Culturally tailored interventions focused on diet, exercise, and stress management may improve glutamate metabolism and fatty acid profiles in Black caregivers, potentially lowering their risk of obesity and hypertension. Finally, expanded health assessment of Black caregivers should be considered, as the effects of chronic stress on processes such as glutamate metabolism may increase their risk for metabolic syndrome.

Supplementary Material

Supplemental Digital Content (SDC) 1
Supplemental Digital Content (SDC) 2
Supplemental Digital Content (SDC) 3
Supplemental Digital Content (SDC) 4

Acknowledgement:

Research reported in this publication was supported by the National Institutes of Health under Award Number P30NR018090 (PI: McCauley), R01AG054079 (PI: Hepburn, Griffiths) and K23AG070378 (PI: Brewster). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Footnotes

The authors have no conflicts of interest to report.

Ethical Conduct of Research: This study was approved by Emory University Institutional Review Board (IRB00103619).

Contributor Information

Glenna S. Brewster, Emory University Nell Hodgson Woodruff School of Nursing, Atlanta, GA, USA.

Madelyn C. Houser, Emory University Nell Hodgson Woodruff School of Nursing, Atlanta, GA, USA.

Irene Yang, Emory University Nell Hodgson Woodruff School of Nursing, Atlanta, GA, USA.

Jordan Pelkmans, Emory University Nell Hodgson Woodruff School of Nursing, Atlanta, GA, USA.

Melinda Higgins, Emory University Nell Hodgson Woodruff School of Nursing, Atlanta, GA, USA.

Cristy Tower-Gilchrist, Emory University Nell Hodgson Woodruff School of Nursing, Atlanta, GA, USA.

Jessica Wells, Emory University Nell Hodgson Woodruff School of Nursing, Atlanta, GA, USA.

Arshed A. Quyyumi, Emory University School of Medicine, Atlanta, GA, USA.

Dean Jones, Emory University School of Medicine, Atlanta, GA, USA.

Sandra Dunbar, Emory University Nell Hodgson Woodruff School of Nursing, Atlanta, GA, USA.

Nicole Carlson, Emory University Nell Hodgson Woodruff School of Nursing, Atlanta, GA, USA.

References

  1. Alzheimer’s Association. (2024). 2024 Alzheimer’s disease facts and figures. Alzheimer’s & Dementia, 20, 3708–3821. 10.1002/alz.13809 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arjmand B, Dehghanbanadaki H, Yoosefi M, Rezaei N, Mohammadi Fateh S, Ghodssi-Ghassemabadi R, Najjar N, Hosseinkhani S, Tayanloo-Beik A, Adibi H, Farzadfar F, Larijani B, & Razi F (2023). Association of plasma acylcarnitines and amino acids with hypertension: A nationwide metabolomics study. PLoS ONE, 18, e0279835. 10.1371/journal.pone.0279835 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Baik D, Centi S, & McNair B (2023). Assessing racial and ethnic differences in cardiovascular disease in U.S. family caregivers of persons with dementia: Analysis of data from the 2015–2020 Behavioral Risk Factor Surveillance System. Research in Gerontological Nursing, 16, 241–249. 10.3928/19404921-20230706-01 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bland V, & Sharma M (2017). Physical activity interventions in African American women: A systematic review. Health Promotion Perspectives, 7, 52–59. 10.15171/hpp.2017.11 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bockus LB, Biggs ML, Lai HTM, de Olivera Otto MC, Fretts AM, McKnight B, Sotoodehnia N, King IB, Song X, Siscovick DS, Mozaffarian D, & Lemaitre RN (2021). Assessment of plasma phospholipid very-long-chain saturated fatty acid levels and healthy aging. JAMA Network Open, 4, e2120616. 10.1001/jamanetworkopen.2021.20616 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Calderón-Pérez L, Gosalbes MJ, Yuste S, Valls RM, Pedret A, Llauradó E, Jimenez-Hernandez N, Artacho A, Pla-Pagà L, Companys J, Ludwig I, Romero M-P, Rubió L, & Solà R (2020). Gut metagenomic and short chain fatty acids signature in hypertension: A cross-sectional study. Scientific Reports, 10, 6436. 10.1038/s41598-020-63475-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Carson AP, Howard G, Burke GL, Shea S, Levitan EB, & Muntner P (2011). Ethnic differences in hypertension incidence among middle-aged and older adults: The multi-ethnic study of atherosclerosis. Hypertension, 57, 1101–1107. 10.1161/HYPERTENSIONAHA.110.168005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Champely S (2017). pwr: Basic functions for power analysis. Retrieved from https://CRAN.R-project.org/package=pwr
  9. Charlson ME, Pompei P, Ales KL, & MacKenzie CR (1987). A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. Journal of Chronic Disorders, 40, 373–383. 10.1016/0021-9681(87)90171-8 [DOI] [PubMed] [Google Scholar]
  10. Chen H-H, Tseng YJ, Wang S-Y, Tsai Y-S, Chang C-S, Kuo T-C, Yao W-J, Shieh C-C, Wu C-H, & Kuo P-H (2015). The metabolome profiling and pathway analysis in metabolic healthy and abnormal obesity. International Journal of Obesity, 39, 1241–1248. 10.1038/ijo.2015.65 [DOI] [PubMed] [Google Scholar]
  11. Cohen J (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Routledge. 10.4324/9780203771587 [DOI] [Google Scholar]
  12. Cookson TA (2021). Bacterial-induced blood pressure reduction: Mechanisms for the treatment of hypertension via the gut. Frontiers in Cardiovascular Medicine, 8, 721393. 10.3389/fcvm.2021.721393 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. de Mello AH, Costa AB, Engel JDG, & Rezin GT (2018). Mitochondrial dysfunction in obesity. Life Sciences, 192, 26–32. 10.1016/j.lfs.2017.11.019 [DOI] [PubMed] [Google Scholar]
  14. Dumas SJ, Bru-Mercier G, Courboulin A, Quatredeniers M, Rücker-Martin C, Antigny F, Nakhleh MK, Ranchoux B, Gouadon E, Vinhas M-C, Vocelle M, Raymond N, Dorfmüller P, Fadel E, Perros F, Humbert M, & Cohen-Kaminsky S (2018). NMDA-type glutamate receptor activation promotes vascular remodeling and pulmonary arterial hypertension. Circulation, 137, 2371–2389. 10.1161/CIRCULATIONAHA.117.029930 [DOI] [PubMed] [Google Scholar]
  15. Fayed N, Lopez Del Hoyo Y, Andres E, Serrano-Blanco A, Bellón J, Aguilar K, Cebolla A, & Garcia-Campayo J (2013). Brain changes in long-term zen meditators using proton magnetic resonance spectroscopy and diffusion tensor imaging: A controlled study. PLoS ONE, 8, e58476. 10.1371/journal.pone.0058476 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Gabor A, & Leenen FH (2012). Central neuromodulatory pathways regulating sympathetic activity in hypertension. Journal of Applied Physiology, 113, 1294–1303. 10.1152/japplphysiol.00553.2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. García-Martín V, de Hoyos-Alonso MC, Delgado-Puebla R, Ariza-Cardiel G, & del Cura-González I (2023). Burden in caregivers of primary care patients with dementia: Influence of neuropsychiatric symptoms according to disease stage (NeDEM project). BMC Geriatrics, 23, 525. 10.1186/s12877-023-04234-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hinnouho G-M, Czernichow S, Dugravot A, Nabi H, Brunner EJ, Kivimaki M, & Singh-Manoux A (2015). Metabolically healthy obesity and the risk of cardiovascular disease and type 2 diabetes: The Whitehall II cohort study. European Heart Journal, 36, 551–559. 10.1093/eurheartj/ehu123 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Horner-Johnson W, Dobbertin K, Kulkarni-Rajasekhara S, Beilstein-Wedel E, & Andresen EM (2015). Food insecurity, hunger, and obesity among informal caregivers. Preventing Chronic Disease, 12, 150129. 10.5888/pcd12.150129 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kim Y-K, & Jeon SW (2018). Neuroinflammation and the immune-kynurenine pathway in anxiety disorders. Current Neuropharmacology, 16, 574–581. 10.2174/1570159X15666170913110426 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Knight BG, Flynn Longmire CV, Dave J, Kim JH, & David S (2007). Mental health and physical health of family caregivers for persons with dementia: A comparison of African American and white caregivers. Aging & Mental Health, 11, 538–546. 10.1080/13607860601086561 [DOI] [PubMed] [Google Scholar]
  22. Kramer EB, Knight EL, & Bryan AD (2023). Cooking interventions for improving diet quality among Black Americans: A randomized controlled trial. Annals of Behavioral Medicine, 57, 323–333. 10.1093/abm/kaac058 [DOI] [PubMed] [Google Scholar]
  23. Kris-Etherton PM, Griel AE, Psota TL, Gebauer SK, Zhang J, & Etherton TD (2005). Dietary stearic acid and risk of cardiovascular disease: Intake, sources, digestion, and absorption. Lipids, 40, 1193–1200. 10.1007/s11745-005-1485-y [DOI] [PubMed] [Google Scholar]
  24. Landi F, Calvani R, Picca A, Tosato M, Martone AM, Ortolani E, Sisto A, D’Angelo E, Serafini E, Desideri G, Fuga MT, & Marzetti E (2018). Body mass index is strongly associated with hypertension: Results from the longevity check-up 7+ study. Nutrients, 10, 1976. 10.3390/nu10121976 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Li S, Park Y, Duraisingham S, Strobel FH, Khan N, Soltow QA, Jones DP, & Pulendran B (2013). Predicting network activity from high throughput metabolomics. PLoS Computational Biology, 9, e1003123. 10.1371/journal.pcbi.1003123 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lin Y-T, Salihovic S, Fall T, Hammar U, Ingelsson E, Ärnlöv J, Lind L, & Sundström J (2020). Global plasma metabolomics to identify potential biomarkers of blood pressure progression. Arteriosclerosis, Thrombosis, and Vascular Biology, 40, e227–e237. 10.1161/atvbaha.120.314356 [DOI] [PubMed] [Google Scholar]
  27. Liu KH, Nellis M, Uppal K, Ma C, Tran V, Liang Y, Walker DI, & Jones DP (2020). Reference standardization for quantification and harmonization of large-scale metabolomics. Analytical Chemistry, 92, 8836–8844. 10.1021/acs.analchem.0c00338 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Ma W, Heianza Y, Huang T, Wang T, Sun D, Zheng Y, Hu FB, Rexrode KM, Manson JE, & Qi L (2018). Dietary glutamine, glutamate and mortality: Two large prospective studies in US men and women. International Journal of Epidemiology, 47, 311–320. 10.1093/ije/dyx234 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Miura K, Stamler J, Nakagawa H, Elliott P, Ueshima H, Chan Q, Brown IJ, Tzoulaki I, Saitoh S, Dyer AR, Daviglus ML, Kesteloot H, Okayama A, Curb JD, Rodriguez BL, Elmer PJ, Steffen LM, Robertson C, & Zhao L (2008). Relationship of dietary linoleic acid to blood pressure: The international study of macro-micronutrients and blood pressure study. Hypertension, 52, 408–414. 10.1161/HYPERTENSIONAHA.108.112383 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Onuh JO, & Qiu H (2021). Metabolic profiling and metabolites fingerprints in human hypertension: Discovery and potential. Metabolites, 11, 687. 10.3390/metabo11100687 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Shi M, He J, Li C, Lu X, He WJ, Cao J, Chen J, Chen J-C, Bazzano LA, Li J-X, He H, Gu D, & Kelly TN (2022). Metabolomics study of blood pressure salt-sensitivity and hypertension. Nutrition, Metabolism and Cardiovascular Diseases, 32, 1681–1692. 10.1016/j.numecd.2022.04.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Shi Z, Yuan B, Taylor AW, Dai Y, Pan X, Gill TK, & Wittert GA (2011). Monosodium glutamate is related to a higher increase in blood pressure over 5 years: Findings from the Jiangsu Nutrition Study of Chinese adults. Journal of Hypertension, 29, 846–853. 10.1097/HJH.0b013e328344da8e [DOI] [PubMed] [Google Scholar]
  33. Tahir UA, Katz DH, Avila-Pachecho J, Bick AG, Pampana A, Robbins JM, Yu Z, Chen Z-Z, Benson MD, Cruz DE, Ngo D, Deng S, Shi X, Zheng S, Eisman AS, Farrell L, Hall ME, Correa A, Tracy RP, … Gerszten RE (2022). Whole genome association study of the plasma metabolome identifies metabolites linked to cardiometabolic disease in Black individuals. Nature Communications, 13, 4923. 10.1038/s41467-022-32275-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Terés S, Barceló-Coblijn G, Benet M, Álvarez R, Bressani R, Halver JE, & Escribá PV (2008). Oleic acid content is responsible for the reduction in blood pressure induced by olive oil. Proceedings from the National Academy of Sciences, 105, 13811–13816. 10.1073/pnas.0807500105 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Tilves C, Yeh H-C, Maruthur N, Juraschek SP, Miller E, White K, Appel LJ, & Mueller NT (2022). Increases in circulating and fecal butyrate are associated with reduced blood pressure and hypertension: Results from the SPIRIT trial. Journal of the American Heart Association, 11, e024763. https://doi.org/doi: 10.1161/JAHA.121.024763 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Torres N, Tobón-Cornejo S, Velazquez-Villegas LA, Noriega LG, Alemán-Escondrillas G, & Tovar AR (2023). Amino acid catabolism: An overlooked area of metabolism. Nutrients, 15, 3378. 10.3390/nu15153378 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Uppal K, Walker DI, & Jones DP (2017). xMSannotator: An R package for network-based annotation of high-resolution metabolomics data. Analytical Chemistry, 89, 1063–1067. 10.1021/acs.analchem.6b01214 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Wang XR, Robinson KM, & Carter-Harris L (2014). Prevalence of chronic illnesses and characteristics of chronically ill informal caregivers of persons with dementia. Age and Ageing, 43, 137–141. 10.1093/ageing/aft142 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Wei D, González-Marrachelli V, Melgarejo JD, Liao C-T, Hu A, Janssens S, Verhamme P, Van Aelst L, Vanassche T, Redon J, Tellez-Plaza M, Martin-Escudero JC, Monleon D, & Zhang Z-Y (2023). Cardiovascular risk of metabolically healthy obesity in two European populations: Prevention potential from a metabolomic study. Cardiovascular Diabetology, 22, 82. 10.1186/s12933-023-01815-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Yelamanchi SD, Jayaram S, Thomas JK, Gundimeda S, Khan AA, Singhal A, Keshava Prasad TS, Pandey A, Somani BL, & Gowda H (2016). A pathway map of glutamate metabolism. Journal of Cell Communication and Signaling, 10, 69–75. 10.1007/s12079-015-0315-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zhang Y, Yu X-J, Chen W-S, Gao H-L, Liu K-L, Shi X-L, Fan X-Y, Jia L-L, Cui W, Zhu G-Q, Liu J-J, & Kang Y-M (2016). Exercise training attenuates renovascular hypertension partly via RAS- ROS- glutamate pathway in the hypothalamic paraventricular nucleus. Scientific Reports, 6, 37467. 10.1038/srep37467 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Zheng Y, Hu FB, Ruiz-Canela M, Clish CB, Dennis C, Salas-Salvado J, Hruby A, Liang L, Toledo E, Corella D, Ros E, Fitó M, Gómez-Gracia E, Arós F, Fiol M, Lapetra J, Serra-Majem L, Estruch R, & Martínez-González MA (2016). Metabolites of glutamate metabolism are associated with incident cardiovascular events in the PREDIMED PREvención con DIeta MEDiterránea (PREDIMED) Trial. Journal of the American Heart Association, 5, e003755. https://doi.org/doi: 10.1161/JAHA.116.003755 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Digital Content (SDC) 1
NIHMS2023197-supplement-Supplemental_Digital_Content__SDC__1.xlsx (16.6KB, xlsx)
Supplemental Digital Content (SDC) 2
NIHMS2023197-supplement-Supplemental_Digital_Content__SDC__2.xlsx (109KB, xlsx)
Supplemental Digital Content (SDC) 3
NIHMS2023197-supplement-Supplemental_Digital_Content__SDC__3.xlsx (117.8KB, xlsx)
Supplemental Digital Content (SDC) 4
NIHMS2023197-supplement-Supplemental_Digital_Content__SDC__4.xlsx (13.3KB, xlsx)

RESOURCES