Up-to-date, genetic studies have identified a limited number of loci associated with human longevity by recognizing age at death or survival up to advanced ages as a specific phenotype. Long-lived people are those who have exceeded ≥90 years [1], some of them having overcome or stabilized or avoided deadly diseases such as cancer and atherosclerosis. The abovementioned characteristics of long-lived populations which led to their longer lifespan imply that beyond environmental factors, genes controlling lifespan may play an important role [2].

Atherosclerosis is a disease that begins very early in life, even in the prenatal and infancy period, and becomes clinically evident in approximately the 4th or 5th decade of life [3]. The diseases sharing the atherosclerotic mechanism are mainly Coronary Artery Disease (CAD), Diabetes Mellitus (DM), hypertension, Alzheimer’s Disease (AD) and obesity [4]. Several genes have been linked to the risk of the progression of these diseases, such as cholesteryl ester transfer protein (CETP) [5], Apolipoprotein E (ApoE) [6, 7], Angiotensin-Converting Enzyme (ACE) [8] and adiponectin (ADIPOQ) [9]. Genes that may be involved in cancer pathogenesis include Insulin-like Growth Factor (IGF), p53 and fork-head box O3A (FOXOA3) [10]. In the present study, we focused on 5 Single Nucleotide Polymorphisms (SNPs) of the CETP, ADIPOQ, IGFBP3 and ACE genes and their related proteins.

CETP interacts with lipoproteins in order to exchange cholesterol esters between Triglycerides (TGs), Low-Density Lipoproteins (LDL) and High-Density Lipoproteins (HDL). High levels of CETP may enhance the formation of small, dense LDL and HDL particles, which are atherogenic [11]. Several CETP gene variants such as TaqIB (rs708272) and I405V (rs5882) are associated with reduced CETP mass and HDL cholesterol (HDL-C) [12, 13]. Both TaqIB and I405V variants are the most widely studied and were found to be associated with CAD [5, 14, 15], left main CAD [16] and longevity [17, 18].

The renin-angiotensin-aldosterone system (RAAS) also plays a role in the maintenance of cardiovascular (CV) homeostasis. The ACE is an important gene of the RAAS that has been evaluated in the pathogenesis of hypertension, CAD, heart failure and, recently, longevity [19]. The most common variant of ACE gene, rs1799752, is associated with hypertension [20], heart failure [21] and lifespan variation [19].

The adipose tissue-derived peptide, AdipoQ, is a cytokine [22]. AdipoQ is a determinant of insulin sensitivity that exerts anti-inflammatory and anti-atherogenic effects [22]. The endocrine function of adipose tissue seems to contribute to several metabolic disorders as well as CV Disease (CVD) [23-25]. Human adipocytes appear to have a number of receptors that are sensitive to various factors that influence important systems such as the endocrine, vascular, immune and nervous system [26]. A common variant of the ADIPOQ gene in chromosome 3 and exon 2 and position +45T>G with rs2241766 (+45T>G) was associated with the risk of CVD [27].

The IGF-1 gene seems to be negatively related to age [28]. Circulating IGF binding protein-3 (IGFBP3) binds > 90% of the circulating IGF-I, thereby reducing the levels of free IGF-I and increasing the risk of DM [29]. Furthermore, the IGFBP3 exerts mitogenic and metabolic activities in growth regulation, survival and cell differentiation [30]. The rs2854744 (A-202C) variant of the IGFBP3 gene was found to be associated with circulating IGFBP-3 levels [31]. There is some evidence that the alleles related to higher circulating IGFBP-3 levels are also associated with a higher risk of early-stage cancers [32].

In the present study, we evaluated serum levels of specific proteins as well as variants of their related genes in long-lived families comprising 3 generations and short-lived families comprising 2 generations. It is suggested that some people may have protective genes and proteins in contrast to people with a disease history “cargo”. This is the first time that potential longevity genes were studied in comparison with their serum protein levels in Greek families. A unique characteristic of this study was that samples were collected from all generations of both family groups, even from the oldest old members. The aim of this study was to investigate the differences in gene variants and serum protein levels between the members of long- and short-lived families.

In the present study, in line with a previous one [34], we compared biochemical and genetic markers between members of families with different lifespans.

The FL1 participants were 5 years older than FD1 group. There is no explanation for this and the difference is too small to have any influence on evaluated proteins levels as well as variants of their related genes.

Higher levels of UA were detected in FD1 individuals compared with FD2. UA as the last product of purine metabolism is a biomarker that is related to inflammation [35]. It is suggested that UA can penetrate cell membrane and exert damaging intracellular actions such as oxidation and inflammation [36]. This action could represent a link between UA and CAD. Apart from CAD, elevated UA may play an important role in the onset of hypertension and diabetic complications [37-40]. In the same line, Malik et al. [41] who studied healthy octogenarians determined that higher levels of UA were associated with vascular inflammation. In addition, Yan et al. [42] reported that elevated levels of UA were associated with an increased risk of cancer incidence and mortality, while Beavers et al. [43] suggested that elevations in UA might lead to sarcopenia. Accumulating evidence suggests that hyperuricemia is one of the important factors that may significantly contribute to the development and progression of CVD and chronic kidney disease [44]. Therefore, it follows that in members of families with a history of shorter lifespan, UA levels were similar between the middle-aged [FD1: 58.5 (51.7-66.3) years] and the oldest old individuals of long-lived families [93 (90-96) years]. Although it is premature to draw any definitive conclusions due to the small samples, these findings may suggest that in families with shorter lifespan, the increase in UA could be an additional factor involved in shortening lifespan.

Additionally, serum GL levels were increased in both groups according to age. The relationship between age, CVD and GL levels [45, 46], as well as the association between increased TGs and CVD [11, 47-49] have been previously investigated. We detected the highest TG levels in the FD1 individuals of the EAD Group. This may imply that while they are ageing, TG concentrations increase in members of short-lived families and this may be an additional CVD risk factor in middle-aged individuals of such families.

In the LON Group, HDL-C was not significantly different in the FL2 compared with the FL1 group (the difference was only 1 mg/dl). In contrast, HDL-C was significantly higher in the FL2 compared with the P group, but the number of samples was small.

AdipoQ levels were increased in the oldest individuals of the LON Group, which may imply that higher adiponectin levels act beneficially in very old individuals. Several studies reported higher adiponectin concentrations in centenarians compared with younger individuals [50-52]. Substantiating this finding, Bik et al. [53] described hyperadiponectinemia in centenarians and found an inverse correlation between AdipoQ and Homeostasis Model Assessment for Insulin Resistance (HOMA-IR).

Independently of family’s lifespan history, the FD2 and FL2 individuals with CETPB2B2 genotype, compared with individuals with CETPB1B1 genotypes, had lower serum CETP levels (p = 0.05, p = 0.03, respectively). However, after adjustment for age and gender, only individuals from the LON group, and more specifically, P and FL2 individuals, had this association. This suggests the possibility of genetic determination in these families and that the B1 allele is unfavourable for long life. Furthermore, low CETP concentrations may act protectively in conjunction with the B2 allele.

It is well-established that the B2 homozygotes have less serum CETP activity or mass than B1 homozygotes (see review by Boekholdt and Thompson [12]). However, we found this correlation only in the families with longevity.

In the present study, Greek individuals of both families with the DD genotype had higher serum ACE levels compared with the II genotype of ACE I/D polymorphism. Yang et al. [54] reported that in Taiwanese individuals with AD, the DD genotype was related to increased levels of ACE in the plasma and that the I allele was associated with a decreased risk of AD; they thus characterized the D allele as a “risk” allele. Several studies linked the increased ACE concentrations with the DD genotype, CAD [55], hypertension [56], autoimmune diseases [57, 58] and acute respiratory distress syndrome [59]. Moreover, Zhang et al. [60] reported that the DD genotype was more frequent in patients with major adverse CV events among CAD patients. On the other hand, there are studies that did not find any association of ACE I/D polymorphism and serum ACE levels in septic patients [61], age-related muscular degeneration [62] and patients with venous thromboembolism [63]. Multivariate regression analysis of gene polymorphisms and their related proteins after adjustment for age and gender also showed that ACE concentrations were higher in individuals with the DD genotype compared with the II genotype in both families.

Human lifespan depends mainly on 2 factors: 1) expression of genes and epigenetics; both are responsible for plasma proteins, enzymes and molecule levels, and, 2) environmental effects, which can influence these variables. Therefore, we evaluated families with longevity or early deaths, so that all members of the same family had potentially similar lifestyle behaviour. The family environment has a critical role in the development of cardiometabolic disorders (such as smoking behaviour, eating habits, obesity and hypertension) in offspring and their children [64]. Additionally, the association of support from family for adoption of healthy eating habits and performing exercise with improvements of self-leadership (defined as a process of behavioural and cognitive self-evaluation and self-influence, in which an individual achieves the self-direction and self-motivation needed to make positive changes in behaviours) is also important as shown in cancer patients [65]. Thus, the parental lifespan history, biochemical phenotype and certain genes could be used as a practical approach for the early preventive measures and identification of families at risk for early death.

The main limitation of this study is the relatively small sample size. However, long-lived individuals represent a very selective group and it is therefore difficult to collect samples from these families. In the present study, samples were collected from all generations of both Family Groups, even from the oldest old members. Furthermore, it is the first time that Greek families both with a history of early death and with long lifespan were evaluated for potential longevity genes and biomarkers.

Increase serum TGs, UA and GL concentrations were higher in the middle-aged individuals compared with their children in families independently of their lifespan. Serum adiponectin concentration was the highest in the oldest old individuals implying beneficial influence on lifespan. Independently of family’s lifespan history, the youngest individuals with CETPB2B2 genotype, compared with individuals with CETPB1B1 genotypes, had lower serum CETP concentrations. The knowledge of the unfavourable gene(s) influencing human lifespan may be helpful in encouraging individuals to follow healthier lifestyle habits and better control their high-risk biomarkers.