Age-associated changes in circulatory fatty acids: new insights on adults and long-lived individuals

Abstract

Long-lived individuals (LLIs) are considered an ideal model to study healthy human aging. Blood fatty acid (FA) profile of a cohort of LLIs (90–111 years old, n = 49) from Sicily was compared to adults (18–64 years old, n = 69) and older adults (65–89 years old, n = 54) from the same area. Genetic variants in key enzymes related to FA biosynthesis and metabolism were also genotyped to investigate a potential genetic predisposition in determining the FA profile. Gas chromatography was employed to determine the FA profile, and genotyping was performed using high-resolution melt (HRM) analysis. Blood levels of total polyunsaturated FA (PUFA) and total trans-FA decreased with age, while the levels of saturated FA (SFA) remained unchanged. Interestingly, distinctively higher circulatory levels of monounsaturated FA (MUFA) in LLIs compared to adults and older adults were observed. In addition, among LLIs, rs174537 in the FA desaturase 1/2 (FADS1/2) gene was associated with linoleic acid (LA, 18:2n-6) and docosatetraenoic acid (DTA, 22:4n-6) levels, and the rs953413 in the elongase of very long FA 2 (ELOVL2) was associated with DTA levels. We further observed that rs174579 and rs174626 genotypes in FADS1/2 significantly affect delta-6 desaturase (D6D) activity. In conclusion, our results suggest that the LLIs have a different FA profile characterized by high MUFA content, which indicates reduced peroxidation while maintaining membrane fluidity.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11357-022-00696-z.

Introduction

Human longevity is a complex trait determined by numerous interacting factors, including genetic, metabolic, environmental, and behavioral characteristics [1–4]. Demographic evidence shows a continuing rise in the age at death, which corresponds to a gradual increase in human longevity [5]. From 2015 to 2050, the percentage of the worldwide population aged over 60 years is estimated to double from 12 to 22% [6]. At the same time, increasing age is also the leading risk factor for major chronic diseases, such as cardiovascular diseases (CVD) and neurodegenerative disorders. However, according to the classic concept of Rowe and Kahn, successful aging is defined as high physical, psychological, and social functioning in old age without major diseases [7, 8]. Exploring the factors that contribute to a long and healthy life is of great importance in order to improve the quality of life of older people [1, 9].

Long-lived individuals (LLIs), i.e., people living 90 years and more, are considered the best model to study healthy human aging. Contrary to what is seen in normal aging, characterized by the disruption of homeostatic processes that might predispose people to major chronic diseases, LLIs have a lower incidence or a higher possibility of escaping, the age-related diseases [10–13]. Investigating the positive phenotypes of LLIs could allow us to modify the aging rate by providing important information on how to slow the aging process.

Previous studies have shown that some cohorts of Sicilian populations, particularly those living in villages located in the Madonie municipalities, have a high number of LLIs compared to other areas in Sicily and Italy [14]. Therefore, this population is ideal for studies aimed at understanding the factors that may determine longevity.

Among aging biomarkers, some fatty acids (FA) have been extensively studied as implicated in aging process development because their profile is critical in maintaining cell and tissue homeostasis [4]. FA not only affect membrane properties, but they also exert receptor-mediated effects through their metabolites [15]. Since the human body cannot synthesize some FA, they must be taken from the diet. These FA are called essential FA (EFA), which are linoleic acid (LA, 18:2n-6) of the n-6 series and the alpha-linolenic acid (ALA, 18:3n-3) of the n-3 series. ALA serves as the precursor for the production of long-chain n-3 PUFA, such as eicosapentaenoic acid (EPA, 20:5n-3), docosapentaenoic acid (DPA, 22:5n-3), and docosahexaenoic acid (DHA, 22:6n-3), whereas LA serves as the precursor of long-chain n-6 PUFA, namely gamma-linolenic acid (GLA, 18:3n-6), dihomo-gamma-linolenic acid (DGLA, 20:3n-6), arachidonic acid (AA, 20:4n-6), and docosatetraenoic acid (DTA, 22:4n-6) through a series of desaturation and elongation processes [16, 17]. Other FA, such as saturated FA (SFA) and monounsaturated FA (MUFA), are also considered nonessential as our body can synthesize them without receiving them directly from the diet [15]. Therefore, the diet is described to have a limited influence on FA status. Indeed, genetic variants that influence the desaturation and elongation of long-chain PUFA may modulate circulating PUFA levels. Metabolic pathways, physio-pathological disorders, and gut microbiota are other factors that affect the FA profile [18].

Among FA, PUFA have been extensively studied and implicated as potential lipid biomarkers of aging [4, 19]. The n-3 and n-6 PUFA and their metabolites affect metabolic, inflammatory, and oxidative processes [20]. The n-6 PUFA-derived lipid mediators have been associated with pro-inflammatory processes, whereas those of n-3 are described to have anti-inflammatory effects [20]. It is well demonstrated that n-3 PUFA, particularly EPA and DHA, are associated with decreased risk of CVD, type 2 diabetes, metabolic syndrome, breast cancer, and depression in different populations [6, 15, 21–25]. Higher n-3 PUFA levels of red blood cells are associated with a reduced risk of all-cause mortality [26, 27]. The intake of n-3 PUFA is also inversely related to the rate of telomere shortening, an important marker of cell aging [28–30]. In addition, lower levels of EPA and DHA are linked with an increased risk of cognitive impairment and dementia, making them a potential marker of brain aging [19, 21, 31, 32]. It has also been suggested that circulating n-6 PUFA levels benefit cardiometabolic outcomes, although controversial findings remained [33–36]. On the contrary, SFA and trans-FA are related to obesity, type 2 diabetes, increased CVD risk, and age-related dementia [37].

Several studies have previously demonstrated that genetic variants in FA desaturase 1/2 (FADS1/2) and elongase of very long FA 2 (ELOVL2) are correlated with circulatory PUFA concentrations in various populations [16, 38]. FADS1/2 genes encode delta-5 (D5D) and delta-6 (D6D) desaturases, respectively [17]. These enzymes are the main determinants of PUFA levels because they catalyze the formation of long-chain PUFA from dietary ALA and LA [17]. Likewise, ELOVL2 encodes one of the key enzymes in the elongation reaction of long-chain PUFA from their precursor [39]. Single-nucleotide polymorphisms (SNPs) with the strongest association with PUFA status are rs174537, rs174579, and rs174626 in the FADS1/2 gene cluster and rs953413 in the ELOVL2 gene [16, 38, 40]. These SNPs are statistically correlated either with a variation in the circulatory levels of n-3 and n-6 PUFA or with a change in desaturase activity (evaluated in terms of product-to-precursor ratio). Finally, a recent study identified an intergenic variant (rs529143) that was observed to modify the effect of plasma n-3 PUFA and DHA levels on leukocyte telomere length (LTL) [41]. The rs529143 is located on chromosome 1 in a region that includes multiple phospholipase genes, such as phospholipase A2 group IID (PLA2G2D) and phospholipase A2 group IIF (PLA2G2F) [41]. Together, these studies show that SNPs in genes encoding enzymes related to the metabolism of PUFA contribute to plasma concentrations of FA.

Based on the effects of FA on health and chronic diseases, this study aims to evaluate whether a distinctive blood FA profile may be associated with longevity in LLIs of Western Sicily compared to adults and older adults from the same area. We also analyzed the gender-related variations in blood FA profiles. We further studied genetic variants in key enzymes related to FA biosynthesis and metabolism that could be responsible for the FA profile differences between the groups. Finally, we investigated whether among LLIs these SNPs are associated with PUFA levels and desaturase activities.

Materials and methods

Study design and participant characteristics

Subjects participating in the “discovery of molecular and genetic/epigenetic signatures underlying resistance to age-related diseases and comorbidities (DESIGN, 20157ATSLF)” project were used for the present investigation. Detailed study design and participant recruitment have been previously described [14]. Briefly, a total of 172 subjects from Western Sicily were enrolled. The participants ranged from 18 to 111 years of age and were recruited at the University of Palermo (Italy) between June 2017 and March 2020. The population was divided into three age groups: adults (69 subjects, age range 18–64 years old), older adults (54 subjects, age range 65–89 years old), and LLIs (49 subjects, age range 90–111 years old). The participants were relatively healthy. Subjects were excluded from the study if they had been diagnosed with chronic and acute diseases, such as neoplastic and autoimmune diseases and severe dementia. The subjects participated voluntarily and written informed consent was obtained from all of them. The study protocol was conducted following the Declaration of Helsinki and its amendments. The Ethics Committee of Palermo University Hospital approved the study (Nutrition and Longevity, No. 032017).

Anthropometric measurements were recorded, including body weight and height, which were used to calculate the body mass index (BMI). In addition, data on eating habits were collected by a food frequency questionnaire (FFQ) that has been used in previous studies [42, 43]. Overnight fasting blood samples were collected from all the participants. According to standard procedures, hematochemical and hematological analyses were performed immediately, including serum total cholesterol (TC), triglyceride (TG), high-density lipoprotein (HDL), and low-density lipoprotein (LDL). The samples were kept at − 80 °C until further analyses.

Analysis of blood FA using gas chromatography

Total FA was extracted from samples of whole blood and analyzed using gas chromatography, as previously described [44]. Initially, FA methyl esters (FAME) were prepared through direct transesterification. The transesterification reaction was performed by adding boron trifluoride-methanol (BF3-MeOH) (12% w/v, 1.5 M) (Acros Organics, Geel, Belgium) to the sample and heating the mixture at 100 °C for 60 min. After cooling to 25 °C, n-hexane (CARLO ERBA Reagents, S.r.l., France) was used to extract the FAME. The aliquot was air-dried in the darkness and subsequently re-dissolved in n-hexane for analysis.

Separation of FAME was carried out on a GC-2010 gas chromatograph (Shimadzu, Kyoto, Japan), equipped with an SP®-2560 capillary gas chromatography column (L × I.D. 100 m × 0.25 mm × 0.2 μm) and a flame ionization detector. The initial temperature of 170 °C was increased to 220 °C over 50 min. FAME were identified by comparison with a standard mixture (Nu-Chek-Prep, Elysian, MN, USA). The calibration of the method was produced from average response factor (RF) and linear regression equations. The average RF was calculated as the average calibration factor of external standard batches. Additionally, a regression analysis formed an association equation between the instrument response (peak area) and the concentration of individual FAME in the analyte. The linearity of the calibration curve was assessed by the coefficient of determination (r2). The r2 for the calibration curves of FA were higher than 0.99 in the range of concentrations examined. The data were analyzed using Shimadzu system GC Solutions software, designed for this system. The retention times of the FAME mixture on the gas chromatography column were used to identify FA. FA concentration measured in nanogram per milliliter was then expressed as a percentage of total FA:

$$FA\%=(peakareacorrespondingtotheFA/sumofallthepeakareascorrespondingtothetotalmixtureoftheFA)\times 100$$

The n-3 index, defined as the amount of EPA and DHA content in erythrocytes as a percentage of the total amount of FA, was then calculated. Additionally, the AA/EPA, SFA/MUFA, n-6/n-3, and trans-fat index were determined. Desaturase activity was estimated by calculating the ratio of the product FA to precursor FA, as described previously [40, 45]. Four estimates of desaturase activity were studied: D6D activity by dividing the percent composition of GLA by LA, D5D activity by dividing the percent composition of AA by DGLA, and aggregate desaturase activity (ADA), by dividing the percent composition of EPA by ALA in the n-3 pathway, and AA by LA in the n-6 pathway.

Selection of SNPs and genotyping profile

Based on previous studies, five SNPs were selected for genotyping: rs174579 (C > T), rs174626 (C > T), rs174537 (G > T), rs953413 (G > A), and rs529143 (A > C) [16, 38]. Three of these SNPs (rs174579, rs174626, and rs174537) are located in the FADS1/2 gene cluster. The rs953413 is found on the ELOVL2 gene, while rs529143 is an intergenic polymorphism in a region that includes multiple phospholipase genes such as PLA2G2D and PLA2G2F. The genomic DNA was extracted from peripheral blood leukocytes and purified using E.Z.N.A.® Blood DNA Mini Kit (Omega Bio-tek, Inc., GA, USA), following the manufacturer’s instructions. Genotyping was performed using polymerase chain reaction (PCR)-based high-resolution melting (HRM) assay with CFX96 Touch Real-Time PCR Detection System (Bio-Rad Laboratories, Inc., Singapore). HRM is a post-PCR analysis method that is based on detecting small differences in PCR melting curves [46]. The pairs of PCR primers for amplification of rs174579, rs174626, rs174537, rs953413, and rs529143 and the PCR-HRM conditions are shown in Table S1. Data were analyzed using Bio-Rad CFX Manager and Bio-Rad Precision Melt Analysis systems designed specifically for HRM analysis. To ensure the reliability of the results, duplicate samples and sequencing-verified genotyped samples were included as quality controls (supplementary material).

Statistical analysis

Variables were assessed for normality using the Shapiro-Wilk's test and skewed data were subsequently log-transformed for all analyses. A two-way analysis of variance (ANOVA) test was used to identify whether age, gender, or their interaction affects anthropometric, clinical, and FA measurements. When statistically significant, a post hoc pairwise comparison test using the Bonferroni method was performed. Deviation from Hardy–Weinberg Equilibrium (HWE) was tested for each SNP using a chi-square test. Linear regression models were used to examine the associations between individual blood FA and SNP genotypes among LLIs. These models were adjusted for gender, BMI, and LDL levels. The genotypes of rs174537, rs953413, and rs529143 were coded 0, 1, and 2 reflecting the number of copies of an allele being tested (additive genetic model). A dominant model was used for rs174579, in which we combined heterozygous with minor homozygous subjects. For rs174626, a recessive model was used by combining heterozygous with major homozygous subjects. The analyses were conducted using the R software, version 4.0.3 (R Foundation for Statistical Computing, Vienna, Austria) and the interface RStudio version 1.4.1717 (R studio, PBC, Boston, MA, USA). A two-tailed p-value of ≤ 0.05 was considered to be statistically significant.

Results

Characteristics of the study population

Characteristics of the study population are presented in Table 1. One hundred seventy-two healthy subjects with an age range of 18–111 years were included. Fifty-five percent of the study cohort were women. All participants were from Western and South-Western Sicily. The population was divided into three age groups, from adults to LLIs. Subjects in the older adults age class showed an increase in BMI (28.2 ± 4.6 kg/m²), with no gender effect. Regarding serum lipids, TC was low in LLIs compared to adults and older adults, whereas TG levels were not different among the groups. HDL levels of LLIs were not significantly different from those observed in adults and older adults. However, LLIs showed lower LDL levels than those observed in adults and older adults. Women had higher HDL levels compared to men. However, there was no effect of gender on the serum levels of TC, TG, and LDL.

Table 1.

Characteristics of study participants

Variable	Adults	Older adults	LLIs	Difference by gender
Age range (y)	18–64	65–89	90–111
Men/women	34/35	28/26	15/34
Weight (kg)	71.67 ± 15.11	74.04 ± 13.35	56.15 ± 13.60a,b	< 0.00001*
BMI (kg/m2)	25.17 ± 4.48	28.21 ± 4.60c	25.10 ± 5.43d	0.09
TC (mg/dl)	179.28 ± 35.48	187.47 ± 28.54	163.29 ± 30.59e,b	0.8
HDL (mg/dl)	53.38 ± 14.28	53.94 ± 12.79	49.96 ± 12.78	0.00002*
LDL (mg/dl)	107.43 ± 27.38	112.23 ± 27.41	89.87 ± 24.98f,b	0.06
TG (mg/dl)	92.03 ± 49.41	111.04 ± 43.92	112.86 ± 49.99	0.9

Data are represented as mean ± SD. ^ap ≤ 0.001, LLIs vs adults; ^bp ≤ 0.001, LLIs vs older adults; ^cp ≤ 0.01, older adults vs adults; ^dp ≤ 0.01, LLIs vs older adults; ^ep ≤ 0.05, LLIs vs adults; ^fp ≤ 0.01, LLIs vs adults. LLIs, long-lived individuals; BMI, body mass index; TC, total cholesterol; HDL, high-density lipoprotein; LDL, low-density lipoprotein; TG, triglyceride. *Significant p-value

Blood FA status among the study participants

Total blood FA profiles of the population by age group are presented in Table 2. Figure 1 graphically depicts the age-associated changes in the blood levels of total PUFA, MUFA, and trans-FA. The Fig S1-S5 show changes in the levels of individual FA by the age group.

Table 2.

Blood FA status of the study population by age group

Variable	Adults (n = 69)	Older adults (n = 54)	LLIs (n = 49)	Difference by gender
ALA (%)	0.32 ± 0.15	0.39 ± 0.15a	0.35 ± 0.13	0.1
EPA (%)	0.79 ± 0.66	0.84 ± 0.76	0.64 ± 0.41	0.6
n-3 DPA (%)	0.95 ± 0.33	0.80 ± 0.30a	0.73 ± 0.25b	0.002*
DHA (%)	3.71 ± 1.12	3.01 ± 1.19a	3.22 ± 1.08	0.5
Total n-3 PUFA (%)	5.78 ± 1.77	5.03 ± 2.00c	4.94 ± 1.51d	0.3
n-3 index	5.54 ± 1.71	4.76 ± 1.93a	4.70 ± 1.45d	0.2
LA (%)	23.23 ± 4.18	21.77 ± 3.78	19.66 ± 3.57b,e	0.8
GLA (%)	0.28 ± 0.1	0.33 ± 0.12	0.32 ± 0.16	0.9
EA (%)	0.26 ± 0.04	0.26 ± 0.04	0.27 ± 0.04	0.3
DGLA (%)	1.86 ± 0.45	1.90 ± 0.51	2.27 ± 0.59b,f	0.07
AA (%)	10.07 ± 2.27	9.16 ± 2.10	9.08 ± 1.85d	0.6
DTA (%)	1.05 ± 0.56	0.84 ± 0.39	0.82 ± 0.44d	0.01*
n-6 DPA (%)	0.24 ± 0.10	0.23 ± 0.09	0.32 ± 0.11b,g	0.1
Total n-6 PUFA (%)	36.99 ± 3.35	34.50 ± 4.33 h	32.75 ± 3.25b,e	0.6
AA/EPA	18.38 ± 10.02	16.49 ± 8.84	18.36 ± 8.41	0.6
n-6/n-3	6.99 ± 2.17	7.87 ± 3.07	7.19 ± 2.14	0.3
PLA (%)	1.06 ± 0.48	1.64 ± 0.69 h	2.09 ± 0.85b,e	0.01*
OL (%)	19.01 ± 3.00	21.60 ± 3.03 h	24.05 ± 3.66b,g	0.02*
Eicosenoic acid (%)	0.22 ± 0.04	0.23 ± 0.06	0.24 ± 0.05	0.8
NA (%)	1.98 ± 0.77	1.81 ± 0.67	1.52 ± 0.56b	0.2
Total MUFA (%)	22.27 ± 3.05	25.27 ± 3.19 h	27.90 ± 3.94b,g	0.01*
MC (%)	0.62 ± 0.23	0.87 ± 0.36 h	0.78 ± 0.28i	0.3
PA (%)	22.59 ± 1.56	23.42 ± 2.24	24.20 ± 1.92b	0.7
STA (%)	10.09 ± 1.87	9.50 ± 1.62	8.49 ± 1.47b,f	0.06
Lignoceric acid (%)	1.13 ± 0.61	0.90 ± 0.39	0.51 ± 0.26b,g	0.002*
Total SFA (%)	34.44 ± 2.79	34.69 ± 3.41	34.00 ± 2.49	0.06
SFA/MUFA	1.58 ± 0.28	1.40 ± 0.23 h	1.25 ± 0.25b,f	0.01*
Trans-PLA (%)	0.09 ± 0.02	0.10 ± 0.03	0.09 ± 0.03	0.9
Trans-OL (%)	0.08 ± 0.03	0.09 ± 0.03	0.07 ± 0.02 g	0.02*
Trans-LA (%)	0.27 ± 0.12	0.22 ± 0.09c	0.18 ± 0.10b	0.1
Total trans-FA (%)	0.44 ± 0.13	0.40 ± 0.10	0.34 ± 0.11b,f	0.06
Trans-fat index	0.35 ± 0.13	0.31 ± 0.10	0.25 ± 0.10b,f	0.02*

Data are represented as mean ± SD. ^ap ≤ 0.01, older adults vs adults; ^bp ≤ 0.001, LLIs vs adults; ^cp ≤ 0.05, older adults vs adults; ^dp ≤ 0.05, LLIs vs adults; ^ep ≤ 0.05, LLIs vs older adults; ^fp ≤ 0.01, LLIs vs older adults; ^gp ≤ 0.001, LLIs vs older adults; ^hp ≤ 0.001, older adults vs adults; ⁱp ≤ 0.01, LLIs vs adults. For fatty acid abbreviations, see the text. LLIs, long-lived individuals. *Significant p-value. Fatty acid levels are expressed as percentage of the total fatty acids

Fig. 1.

Mean values ± SD of total n-3 PUFA (A), total n-6 PUFA (B), total MUFA (C), and total trans-FA (D), in 69 adults (18–64 years); 54 older adults (65–89 years) and 49 LLIs (90–111 years). “^a”: p ≤ 0.05, older adults vs adults; “^b” p ≤ 0.05, LLIs vs adults; “^c” p ≤ 0.001, older adults vs adults; “^d” p ≤ 0.001, LLIs vs adults; “^e” p ≤ 0.05, LLIs vs older adults; “^f” p ≤ 0.001, LLIs vs older adults; “^g” p ≤ 0.01, LLIs vs older adults. PUFA, polyunsaturated fatty acids; MUFA, monounsaturated fatty acids; FA, fatty acids; LLIs, long-lived individuals. Fatty acid levels are expressed as percentage of the total fatty acids

Compared with adults, we found that older adults, but not LLIs, have higher levels of ALA. The values of the DHA in LLIs were similar to adults. However, the DHA value decreased in older adults compared to adults. Although not significant, a trend toward a reduction in EPA levels with increasing age was also detected. Total n-3 PUFA decreased with age because of overall decreases in EPA, n-3 DPA, and DHA levels. The reduction in the n-3 index was small but significant in older adults and LLIs with respect to adults.

Total n-6 PUFA decreased with advancing age, primarily due to decreased LA and AA levels. The percentage of LA, the essential n-6 PUFA, significantly differed in LLIs compared to both adults and older adults, with the highest blood levels in adults and the lowest in LLIs. Mean levels of AA (the major n-6 PUFA in erythrocyte membranes) and DTA also decreased in LLIs. However, other LA metabolites, such as DGLA and n-6 docosapentaenoic acid (DPA, 22:5n-6), were significantly higher in LLIs than in adults and older adults. No significant difference in AA/EPA and the n-6/n-3 ratio was observed among the groups.

Total SFA compositions remained stable across the lifespan, and there were no gender differences. However, overlapping values and variations in the individual SFA were observed. In particular, the levels of myristic acid (MC, 14:0) and palmitic acid (PA, 16:0) were higher in LLIs compared to adults and older adults, even though the latter was not statistically significant. In contrast, stearic acid (STA, 18:0) and lignoceric acid (24:0) decreased with advancing age.

We also found significant differences in between-group comparisons of total MUFA, with LLIs and adults showing the highest and lowest values, respectively. Consequently, the SFA/MUFA ratio decreased in LLIs compared to adults and older adults. Circulatory levels of the PA metabolite, palmitoleic acid (PLA, 16:1n-7), were significantly different among the groups, with the highest values in LLIs compared to adult and older adult subjects. Likewise, oleic acid (OL, 18:1n-9) increased with age, while similar blood levels of eicosenoic acid (20:1n-9) were observed among the study groups.

Regarding trans-FA, the FA profile from the LLIs showed a smaller percentage of trans-LA (18:2n-6t), total trans-FA, and trans-fat index compared with both adult and older adult groups. Furthermore, the FA profile from LLIs showed a reduced proportion of trans-OL (18:1t) compared to adults.

Men and women were significantly different in n-3 DPA, DTA, PLA, OL, lignoceric acid, trans-OL, total MUFA, SFA/MUFA ratio, and trans-fat index. Women had lower n-3 DPA, DTA, lignoceric acid, trans-OL, trans-fat index, and SFA/MUFA values, while men had lower PLA, OL, and total MUFA percentages. As a result of these differences, we included sex as a covariate in the subsequent analyses. In addition, we observed a significant interaction between age and gender with regard to n-3 DPA, LA, AA, and DTA (p ≤ 0.05).

We further estimated the D6D, D5D, and ADA activities reflected in n-3 and n-6 PUFA among the three age classes (see Table 3). LLIs had a higher GLA concentration to LA levels than adults, which serves as an indirect marker of D6D enzyme activity. On the other hand, LLIs showed lower D5D enzyme activity than adults and older age groups, as estimated from AA to DGLA concentration levels. No effect of gender was observed.

Table 3.

Estimates of D6D, D5D, and ADA reflected in n-3 and n-6 PUFA of the study population by age group

Variable	Adults (n = 69)	Older adults (n = 54)	LLIs (n = 49)	Difference by gender
D6D (GLA/LA)	0.010 ± 0.007	0.015 ± 0.006 a	0.020 ± 0.010b	0.7
D5D (AA/DGLA)	5.77 ± 1.95	5.09 ± 1.54	4.28 ± 1.47c,b	0.07
n-3 ADA (EPA/ALA)	3.20 ± 4.09	2.71 ± 3.64	2.00 ± 1.49	0.5
n-6 ADA (AA/LA)	0.46 ± 0.17	0.44 ± 0.15	0.49 ± 0.18	0.8

Data are represented as mean ± SD. ^ap ≤ 0.01, older adults vs adults; ^bp ≤ 0.001, LLIs vs adults; ^cp ≤ 0.05, LLIs vs older adults. For fatty acid abbreviations, see the text. LLIs, long-lived individuals; D6D, delta-6 desaturase; D5D, delta-5 desaturase; ADA, aggregate desaturase activity

Characterization of PUFA status among LLIs

The characterization of the PUFA status in LLIs showed that there were no systematic differences between men and women, except for GLA (see Table 4). The average n-3 index was 4.7 ± 1.45. Around 63% of the participants had an n-3 index ranging between 4 and 8%, while 33% had an n-3 index < 4%. As for the AA/EPA ratio, the average value was 18.36 ± 8.41. Nearly 40% of the LLIs had an AA/EPA ratio of less than 15, while the remaining participants had an AA/EPA ratio higher than 15.

Table 4.

Blood PUFA status among LLIs

	Total LLI population (n = 49)	Women (n = 34)	Men (n = 15)
ALA (%) EPA (%) n-3 DPA (%) DHA (%) LA (%) GLA (%)* EA (%) DGLA (%) AA (%) DTA (%) n-6 DPA (%) n-3 index AA/EPA n-6/n-3	0.35 ± 0.13 0.64 ± 0.41 0.73 ± 0.25 3.22 ± 1.08 19.66 ± 3.57 0.32 ± 0.16 0.27 ± 0.04 2.27 ± 0.59 9.08 ± 1.85 0.82 ± 0.44 0.32 ± 0.11 4.70 ± 1.45 18.36 ± 8.41 7.19 ± 2.14	0.37 ± 0.13 0.61 ± 0.42 0.75 ± 0.27 3.25 ± 1.20 19.04 ± 3.06 0.30 ± 0.17 0.27 ± 0.04 2.27 ± 0.62 9.36 ± 1.89 0.84 ± 0.48 0.33 ± 0.10 4.77 ± 1.60 18.99 ± 8.87 7.12 ± 2.30	0.32 ± 0.12 0.59 ± 0.24 0.69 ± 0.19 3.15 ± 0.77 21.08 ± 4.31 0.37 ± 0.14 0.27 ± 0.05 2.27 ± 0.53 8.46 ± 1.65 0.75 ± 0.35 0.29 ± 0.11 4.56 ± 1.05 16.93 ± 7.36 7.36 ± 1.77

Data are represented as mean ± SD. LLIs, long-lived individuals. For fatty acid abbreviations, see the text. *p = 0.03. Fatty acid levels are expressed as percentage of the total fatty acids

Genotype and allele frequencies of the examined SNPs

The genotypes of the SNPs in the FADS1/2 gene cluster (rs174579, rs174626, and rs174537), ELOVL2 (rs953413), and the intergenic rs529143 were determined in 98% (n = 168) of the participants (see Table 5). The genotype distribution for each of the examined SNPs was consistent with HWE (p > 0.05). The minor allele frequency (MAF) of the SNPs ranged between 21 and 49%. No significant differences in the allelic frequencies were observed between the age groups.

Table 5.

Allele and genotype frequency of rs174579, rs174626, rs174537, rs953413, and rs529143 polymorphisms among adults, older adults, and LLIs

SNP	Gene (locus)		Genotype and allele frequency						Chi-square (p-value)
			Adults		Older adults		LLIs		Adults vs LLIs	Older adults vs LLIs
			N	%	N	%	N	%
rs174579	FADS1/2 (chr11:61,838,141)	CC genotype	43	62	33	62	25	55
		CT genotype	23	34	16	30	19	41
		TT genotype	3	4	4	8	2	4
		MAF %		21		23		25	0.30 (0.6)	0.05 (0.8)
rs174626	FADS1/2 (chr11:61,869,585)	TT genotype	18	26	9	17	11	24
		CT genotype	35	51	27	51	23	50
		CC genotype	16	23	17	32	12	26
		MAF %		49		42		49	0.06 (0.8)	1.15 (0.3)
rs174537	FADS1/2 (chr11:61,785,208)	TT genotype	8	12	5	9	5	11
		GT genotype	27	39	28	53	23	50
		GG genotype	34	49	20	38	18	39
		MAF %		31		36		36	0.4 (0.5)	< 0.01 (0.99)
rs953413	ELOVL2 (chr6:11,012,626)	AA genotype	15	22	11	21	8	17
		AG genotype	32	46	25	47	28	61
		GG genotype	22	32	17	74	10	22
		MAF %		45		44		48	0.09 (0.8)	0.12 (0.7)
rs529143	Intergenic (chr1:20,125,527)	CC genotype	5	7	4	8	6	13
		AC genotype	31	45	22	41	18	39
		AA genotype	33	48	27	51	22	48
		MAF %		30		28		33	0.10 (0.7)	0.25 (0.6)

SNP, single-nucleotide polymorphism; LLIs, long-lived individuals; FADS, fatty acid desaturase; MAF, minor allele frequency; ELOVL2, elongase of very long fatty acids 2

Associations of SNP genotypes with PUFA status and desaturase enzyme activity in LLIs

It was investigated whether the polymorphisms examined could alter blood PUFA levels and the FADS indices in LLIs (see Table 6 and Table 7). A positive correlation was identified between rs174537 genotypes and blood levels of LA, eicosadienoic acid (EA, 20:2n-6), and DTA. The major allele carriers of rs174537 had significantly higher levels of DTA, but lower levels of LA and EA compared with the minor allele carriers. Similarly, the genotypes of rs953413 were associated with the DTA, where the presence of the minor allele (A) was associated with lower percentages of DTA. Regarding rs529143, levels of n-3 PUFA were observed. Although the major allele carriers had higher levels of all the observed n-3 PUFA, no significant correlation was observed.

Table 6.

Associations of blood PUFA levels with rs174537, rs953413, and rs529143 genotypes among LLIs

	Genotypes				Adjusted R-square (p-value)
rs174537	G/G (n = 18)	G/T (n = 23)		T/T (n = 5)
ALA (%)	0.40 ± 0.11	0.35 ± 0.15		0.33 ± 0.09	0.1 (0.2)
EPA (%)	0.73 ± 0.44	0.64 ± 0.41		0.40 ± 0.24	0.04 (0.3)
n-3 DPA (%)	0.74 ± 0.24	0.69 ± 0.20		0.60 ± 0.15	0.11 (0.1)
DHA (%)	3.20 ± 0.99	3.29 ± 1.19		2.51 ± 0.47	− 0.08 (0.8)
LA (%)	19.85 ± 2.93	19.20 ± 2.21		23.44 ± 6.60	0.41 (0.0006)*
GLA (%)	0.34 ± 0.22	0.33 ± 0.11		0.31 ± 0.08	0.11 (0.1)
EA (%)	0.25 ± 0.04	0.28 ± 0.04		0.30 ± 0.08	0.27 (0.01)*
DGLA (%)	2.07 ± 0.52	2.41 ± 0.64		2.60 ± 0.46	0.05 (0.3)
AA (%)	9.48 ± 1.80	8.86 ± 1.85		7.77 ± 1.30	0.03 (0.3)
DTA (%)	0.76 ± 0.28	0.73 ± 0.36		0.67 ± 0.08	0.51 (< 0.0001)*
n-6 DPA (%)	0.30 ± 0.08	0.32 ± 0.16		0.27 ± 0.06	0.15 (0.08)
rs953413	G/G (n = 10)	A/G (n = 28)		A/A (n = 8)
ALA (%)	0.35 ± 0.13	0.35 ± 0.12		0.43 ± 0.14	0.14 (0.09)
EPA (%)	0.68 ± 0.52	0.65 ± 0.43		0.60 ± 0.16	0.04 (0.3)
n-3 DPA (%)	0.74 ± 0.28	0.69 ± 0.02		0.69 ± 0.11	0.08 (0.2)
DHA (%)	3.16 ± 1.22	3.26 ± 1.09		2.87 ± 0.81	− 0.07 (0.7)
LA (%)	19.8 ± 2.39	20.43 ± 3.82		18.27 ± 1.96	0.06 (0.2)
GLA (%)	0.29 ± 0.10	0.36 ± 0.19		0.29 ± 0.07	0.1 (0.2)
EA (%)	0.25 ± 0.03	0.29 ± 0.05		0.25 ± 0.04	− 0.01 (0.5)
DGLA (%)	2.03 ± 0.69	2.39 ± 0.55		2.31 ± 0.61	− 0.1 (0.9)
AA (%)	9.21 ± 2.30	8.98 ± 1.86		8.67 ± 1.00	− 0.002 (0.4)
DTA (%)	0.74 ± 0.31	0.74 ± 0.34		0.71 ± 0.19	0.50 (< 0.0001)*
n-6 DPA (%)	0.29 ± 0.12	0.31 ± 0.09		0.31 ± 0.08	0.06 (0.2)
rs529143	A/C + A/A (n = 40)		C/C (n = 6)
ALA (%)	0.37 ± 0.13		0.36 ± 0.10		0.13 (0.09)
EPA (%)	0.65 ± 0.44		0.61 ± 0.15		0.05 (0.2)
n-3 DPA (%)	0.72 ± 0.22		0.60 ± 0.13		0.15 (0.06)
DHA (%)	3.20 ± 1.12		2.99 ± 0.58		− 0.06 (0.7)
Total n-3 PUFA (%)	4.93 ± 1.58		4.57 ± 0.75		− 0.04 (0.6)

Data are represented as mean ± SD. LLIs, long-lived individuals. For fatty acid abbreviations, see the text. *Significant p-value. Fatty acid levels are expressed as percentage of the total fatty acids

Table 7.

Estimates of D6D, D5D, and ADA reflected in n-3 and n-6 PUFA according to rs174579 and rs174626 genotypes among LLIs

	Genotype		Adjusted R-square (p-value)
rs174579	C/C (n = 25)	C/T + T/T (n = 21)
D6D (GLA/LA)	0.018 ± 0.008	0.016 ± 0.006	0.18 (0.04)*
D5D (AA/DGLA)	4.59 ± 1.51	3.71 ± 1.20	0.09 (0.1)
n-3 ADA (EPA/ALA)	2.09 ± 1.38	1.80 ± 1.57	− 0.01 (0.5)
n-6 ADA (AA/LA)	0.50 ± 0.13	0.43 ± 0.13	0.16 (0.07)
rs174626	T/T (n = 11)	C/C + C/T (n = 35)
D6D (GLA/LA)	0.018 ± 0.011	0.017 ± 0.006	0.17 (0.04)*
D5D (AA/DGLA)	5.12 ± 2.00	3.89 ± 1.08	0.08 (0.2)
n-3 ADA (EPA/ALA)	1.66 ± 1.10	2.05 ± 1.56	− 0.04 (0.6)
n-6 ADA (AA/LA)	0.52 ± 0.15	0.45 ± 0.13	0.14 (0.08)

Data are represented as mean ± SD. LLIs, long-lived individuals. D6D, delta-6-desaturase; D5D, delta-5-desaturase; ADA, aggregate desaturase activity. For fatty acid abbreviations, see the text. *Significant p-value

Additionally, desaturase activities were estimated by calculating the product/precursor ratio as indices to evaluate the potential differences in desaturase activities among the LLIs grouped by rs174579 and rs174626 genotypes. The differences were observed for D6D, D5D, n-3 ADA, and n-6 ADA, where the D6D index differed significantly between major and minor allele carriers of both rs174579 and rs174626 after adjustment for covariates (see Table 7). The D6D was lower in the minor allele carriers compared to the major allele carriers for rs174579. The relationships were contrary for rs174626, where the minor allele carriers had a higher D6D index than major allele carriers. Neither D5D, n-3 ADA, nor n-6 ADA was associated with rs174579 genotypes. Likewise, for rs174626 genotypes, no differences in their PUFA product/precursor ratios were found for D5D, n-3 ADA, and n-6 ADA.

Discussion

Over the past decade, various epidemiological studies on genetic, demographic, and phenotypic characteristics of longevity support that LLIs are an ideal model of healthy human aging [10–12]. Despite decreased physical functioning, LLIs have a high ability to adapt to age-associated challenges, and the majority of them endures or escapes diseases that cause death at younger ages [12].

We hypothesized that blood FA, mirroring the whole-body status, could signify physiological decline and better aging. We determined blood FA profile in 69 adults (18–64 years old), 54 older adults (65–89 years old), and 49 LLIs (90–111 years old) and assessed the role of gender in determining the blood FA composition. The results showed differences in FA profile among the three age groups, implying a change in physical properties of the blood cell’s membrane with advancing age.

Gender differences in FA profile

We assessed the role of gender in determining the blood FA composition in the population. Gender is considered a possible confounding factor in studies investigating FA, specifically PUFA status, because sex hormones impact enzymes involved in synthesizing long-chain PUFA and FA metabolism [47]. A systematic review of 51 studies in humans showed that plasma values of AA and DHA are significantly lower in men than women, although there were no gender differences in the levels of their precursors, i.e., LA and ALA [47]. In our study population, women had lower proportions of n-3 DPA, DTA, lignoceric acid, trans-OL, trans-fat index, and SFA/MUFA. In contrast, men had lower PLA, OL, and total MUFA proportions.

Age-associated changes in FA profile

There was also a small but significant reduction in the n-3 index in older adults and LLIs compared to adults. The n-3 index is associated with a lower risk of fatal coronary heart disease (CHD) [48]. The cardioprotective target level for the n-3 index is around 8%, and the level associated with the increased risk for CHD death is < 4% [49]. A systematic review of healthy adults showed high % EPA + DHA values (> 8%) in European countries, such as Denmark and Norway, and moderate n-3 index values (4–6%) in Finland and Sweden. In contrast, very low blood levels were observed in Italy (≤ 4%) [50]. The majority of the LLIs in the present study had an average n-3 index between 4 and 8% (67%, n = 33), while the n-3 index in 33% (n = 16) of LLIs was < 4%. The average moderate value of the n-3 index suggests a decreased risk for chronic diseases among LLIs in this study.

Compared with adults, we found that older adults, but not LLIs, have higher levels of ALA, which is indicative of an accumulation of the precursor of n-3 PUFA metabolic pathways. Despite the high ALA levels, DHA levels were significantly reduced in older adults compared to adults. The reduced DHA levels among older adults might be explained by the age-associated decline in enzyme activity or decreased absorption of DHA. Despite their age, the LLIs maintained DHA levels similar to adults. Similar data have been reported previously in a population from central Italy, in which higher erythrocyte membrane levels of DHA in centenarians were observed compared with elderly (61 to 99 years old) subjects [51]. Moreover, Puca et al. have examined FA profile of erythrocyte membranes as a possible biomarker of longevity by studying another model of successful aging and longevity, i.e., nonagenarian children [4]. They demonstrated a number of modifications of the erythrocyte membrane components including an increase in n-3 PUFA compared to the older population. The results of the genetic analysis did not indicate a role of the genetic loci in modulating the lipid composition observed in nonagenarian offspring erythrocyte membranes. In agreement with this, the conversion rate from ALA to EPA and especially to DHA is described to be low [52, 53]. Thus, it is clear that DHA values are lower in older adults than in adults but not in LLIs and apparently genetics do not play a role. Concerning the possible role of diet to explain the differences between older adults and LLIs, in our survey (Aiello et al., 2021 quoted [14]), we noted no dietary differences between these two populations. Thus, further studies are needed to go insight to this difference between older adults and LLIs.

Regarding EPA, there was a trend in decreasing its amounts with age, but this change did not reach a significant level. Nishihira et al. reported similar results among community-dwelling octogenarians (80 to 94 years old) in Okinawa, in which serum EPA levels did not change significantly with increasing age [54]. In contrast, other studies in different populations have demonstrated that increasing age predicts higher circulatory n-3 PUFA, especially EPA and DHA [55–57]. Increasing EPA and DHA with age in these studies is suggested to be an artifact, resulting from older people consuming more fish and n-3 supplements than younger people [58]. Therefore, it is speculated that aging may not augment n-3 PUFA, but greater n-3 PUFA levels may promote longevity.

Regarding n-6 PUFA, LA levels were inversely associated with age in our study. A similar trend was seen in other studies, where the plasma LA declined with age [59, 60]. The reason for the inverse relations between age and LA levels is not well understood. One possible explanation would be a change in dietary LA intake [58]. While the increase in LA value was reflected by the level of its D5D metabolite AA, other important LA metabolites, such as DGLA, were higher in LLIs than in adults and older adults. DGLA is a product of GLA and a precursor of AA that acts as a substrate for cyclooxygenases and lipoxygenases to produce anti-inflammatory eicosanoids [61]. AA, is a precursor to several pro-inflammatory mediators, such as prostaglandins and leukotrienes. It is either obtained from the diet or endogenously synthesized from LA [62]. Modifying dietary LA does not affect AA circulatory levels, and LA availability is not rate limiting for AA synthesis [58, 63]. Therefore, metabolic factors might determine membrane AA levels [58]. Our result of a reduced AA level with age is consistent with previous evidence, reporting lower levels of both LA and AA in nonagenarian offspring from Southern Italy than in matched controls [4]. The authors suggested that low levels of the n-6 PUFA may speculate a reduced peroxidizability in the cell membrane of nonagenarian offspring.

Surprisingly, our results showed high total MUFA and OL levels among LLIs compared to other age classes. We suppose that the high consumption of the extra virgin olive oil (EVOO) may have influenced this datum. Indeed, in Sicily, EVOO is the primary source of MUFA [64], and we observed that the frequency of EVOO among LLIs was greater than in adults and older adults. MUFA from olive oil was associated with a significant risk reduction of all-cause mortality, cardiovascular mortality, and cardiovascular events [65]. Higher levels of MUFA indicate the maintenance of cell membrane fluidity and resistance toward peroxidation [66]. The high consumption of EVOO (characteristics of the Sicilian diet [64]) also leads to a higher proportion of MUFA in the mitochondrial membrane, which is associated with a better mitochondrial function and a reduction of mitochondrial oxidative stress during aging [67]. Our analysis also showed high PLA levels among LLIs with respect to adults and older adults. This agrees with a recent study by Manca et al. on elderly people from the longevity Blue Zone of Sardinia/Italy, in which elderly people from a high longevity zone of Sardinia showed higher levels of PLA compared with the elderly from a low longevity zone of the same island [18]. The so-called Blue Zone are areas of the world inhabited by exceptionally long-lived populations. The Blue Zone of Sardinia is located in the central-eastern mountain area of the island that displays one of the highest concentrations of LLIs in the world [18]. High levels of circulatory PLA suggest an efficient adipose tissue de novo lipogenesis (DNL), which is associated with decreased lipid accumulation and enhanced insulin sensitivity [68, 69]. DNL is also involved in caloric restriction, which extends lifespan and delays age-related dysfunction [70]. In fact, our LLIs showed a better cholesterol profile than that of the other groups. In contrast, we observed small but significantly higher PA levels among LLIs than in adults. PA is PLA precursor, and elevated levels of this SFA exert detrimental effects on brain cells and might increase the risk of neurodegenerative disorders [71]. However, our LLIs do not show any neurodegenerative disorders at the time of the recruitment.

In addition, we observed lower levels of total trans-FA and trans-fat index in the LLIs compared with adults and older adults. Because elderly people have limited consumption of foods made with processed sources of trans-FA, age-related changes in dietary habits are possible reasons for this trend.

genetic variants and their association with PUFA status

Regarding genetic analyses, previous studies described that people with the T allele of rs174537 polymorphism in FADS1 have lower AA levels and get more benefit from n-3 PUFA than those with the GG allele [16, 72]. The GG genotype of rs174537 is linked with increased PUFA-derived eicosanoid levels that influence downstream inflammatory pathways [73, 74]. The minor allele T of rs174537 is associated with decreased risk of major depressive disorder and breast cancer among women [72, 75]. In addition, the T allele predicts a reduced risk of coronary artery disease (CAD) [76–78]. The rs174537 also modulates the age-associated changes in serum long-chain PUFA, D5D activity, and oxidative stress [79]. Age-associated increases in AA levels and D5D activity have been observed in GG individuals, but not in T allele carriers [79]. Consistent with previous studies, our results confirmed a significant effect of FADS1 rs174537 polymorphism on LA, EA, and DTA levels among LLIs [16]. We also observed that GG carriers have higher ALA, EPA, and AA levels than T allele carriers, but the correlation was insignificant. However, the DHA levels were not linked with rs174537 polymorphism in our study [16].

Besides rs174537, the FA profile is influenced by rs174579 and rs174626 in the FADS1/2 gene cluster. Previous studies consistently reported that the minor allele carriers for rs174579 and rs174626 polymorphisms have lower levels of AA and a reduced desaturase index [38, 40]. This was also observed in the present study, where minor allele carriers for the examined SNPs showed a reduced D6D index (see Table 7). Therefore, minor allele carriers appear to have a reduced capacity to convert LA into GLA. This result confirms the association between variation in the FADS1/2 gene cluster and FA metabolism among LLIs. However, we did not observe any effect of rs174579 and rs174626 for D5D, n-3 ADA, and n-6 ADA in our population [38].

Another SNP we examined was rs953413 found in the locus of ELOVL2 that is involved in the homeostasis of longer chain n-3 PUFA. The minor allele A of rs953413 is described to result in higher EPA but lower DHA levels, whereas the G allele carriers have high DHA levels [80]. The presence of the G allele also shows a protective effect on lipid metabolism and might be an indicator of lower cardiometabolic risk [39, 81]. Although not significant, we observed that AA individuals of this SNP have lower DHA levels. Our result also showed that the minor allele carriers have lower percentages of DTA [16]. However, no relationship between this SNP and EPA level was observed among LLIs. This could be explained by the fact that significant associations are often harder to find with smaller sample sizes [82]. In addition, the genetic background that determines EPA levels in our LLIs population might be different than the populations of previous studies.

Besides FADS and ELOVL2 genes, a recent study on a Singaporean Chinese population showed that an intergenic variant (rs529143) modifies the effects of n-3 PUFA on LTL. Subjects carrying the minor C allele of rs529143 have shorter LTL in the lower tertile of n-3 PUFA and DHA. In comparison, subjects with higher tertile of n-3 PUFA and DHA have longer LTL. Regional genes around rs529143 include multiple phospholipase genes such as PLA2G2D and PLA2G2F that contribute to phospholipid metabolism [41]. Although the major allele carriers of our population had higher levels of all the observed n-3 PUFA, no significant correlation was observed.

Our study provides additional evidence that genetic variants in FADS1/2 and ELOVL2 are associated with levels of certain PUFA and desaturase activity among LLIs. However, since the allele frequency of these SNPs is similar to those observed in adults and older adults, they might have minimal impact on longevity in a population of LLIs from southern Italy.

Limitations and strengths of the study

Our study has certain limitations. First, the number of participants was relatively small. Second, detailed information on the dietary quantity of FA intake and physical activities, which might confound data on FA blood levels, were not evaluated. Moreover, it is possible that other additional factors, unrelated to FA profile but associated with aging, could affect the examined FA. Therefore, studying new factors that may act as confounders will require further investigations in the future. In addition, we used the percentage of product-to-precursor approach to estimate desaturase activity and not their exact concentration. However, previous research has stated that estimates of desaturase activity based on percentage levels of FA are comparable to estimates made using product-to-precursor concentrations [40].

Our study also has strengths. We studied a small but well-characterized group of healthy LLIs, which may provide a better understanding of the factors implicated in the aging process in the presence of well-preserved physiological functions. Moreover, they represent a population very difficult to recruit both to the time and to the budget expenditure related to the home visit.

Conclusion

Our data collected from circulating FA suggest peculiar changes in FA status with advancing age. We observed a reduction in blood levels of total PUFA and trans-FA with age, whereas total SFA remained unchanged throughout life. On the other hand, the FA profile of Sicilian LLIs was characterized by distinctively higher circulatory levels of MUFA compared to adults and older adults. This suggests that the circulatory MUFA content may represent a potential marker of longevity in this population. Additionally, our study provides further evidence that SNPs in FADS1/2 and ELOVL2 affect PUFA status and desaturase indices among LLIs. However, since the allele and genotypic frequency were similar between the age groups, the examined SNPs might have minimal impact on longevity in this population from southern Italy. Finally, while we did not detect any association between the examined SNPs and MUFA levels among LLIs (Table S2), future studies can further investigate the putative link of genetic variants at FADS1/2 and ELOVL2 genes with MUFA levels.

Supplementary Information

Below is the link to the electronic supplementary material.

Funding

This work was funded by the 20157ATSLF project (Discovery of molecular, and genetic/epigenetic signatures underlying resistance to age-related diseases and comorbidities), granted by the Italian Ministry of Education, University, and Research to C.C. and G.C. The Institutional Ethics Committee (“Paolo Giaccone”, University Hospital) approved the study protocol (Nutrition and Longevity, No. 032017). We thank all the donors (or their caregivers) for their kind participation in this study.

Declarations

Conflict of interest

The authors declare no competing interests.