Telomeres are looped DNA (TTAGGG) repeats at the ends of the chromosomes, protecting the chromosome from degradation (30). The length of telomeres reduces progressively with each cell cycle, primarily due to the inability of DNA polymerase to completely replicate the telomeres (31). When the telomeres shorten to a critical length, the cell cycle is arrested, and the cell enters a state of senescence and activates apoptosis (32). The length of the telomere is maintained by an enzyme called telomerase (31). Telomerase enzyme contains two components: Telomerase RNA component (TERC) and telomerase reverse transcriptase (TERT). TERC includes a template for telomeric DNA, which is used by TERT to synthesise new telomeric DNA repeats. Telomerase activity reduces with age (33), leading to short telomeres, and short telomeres have been linked to the progression of cancer, diabetes, and CVDs (34). In a study conducted on TERC⁻/⁻ mice, the catalytic unit of the telomerase enzyme, mice displayed thinning of ventricular walls and cardiomyocyte apoptosis (35). Brouillette et al. conducted a trial with 1,500 participants to investigate the correlation between telomere length and the occurrence of coronary heart disease. They observed that the mean telomere length reduced by around 9% each year and that the participants with short telomere length had a higher risk of developing coronary heart disease than those with comparatively longer telomeres (36). Similarly, coronary endothelial cells with short telomeres exhibit impaired function and formation of plaques leading to age-dependent coronary artery disease (37, 38). Despite several studies indicating the role of short telomeres in the progression of CVDs, a large study including 7,827 participants found telomere length was not significantly associated with cardiovascular mortality (39). The reason for the inconsistency can be multi-factorial such as ethnicity of individuals, medical history, and other factors such as hypertension, smoking, and obesity, as these can attribute to telomere shortening (39, 40). Thus, using telomere length as a marker for age-associated CVD needs more conclusive evidence (41).

Epigenetic changes, including methylation of DNA, RNA methylation, and histone modifications, have been demonstrated to be associated with the development of ageing-induced CVDs (42). DNA methylation involving methylation at 5-cytosine is a highly conserved process in plants, animals, and fungi (43). DNA methylation and demethylation are essential for regulating gene expression, splicing, transposon silencing, and genomic instability (44). DNA methylation at the promoter regions inhibits the binding of transcription factors, which suppresses transcription, whereas demethylation leads to gene expression (45). Studies have identified DNA hypermethylation with age (increased DNA methylation) (44, 46). DNA hypermethylation has been observed in atherosclerotic lesions, as hypermethylation represses the expression of atheroprotective genes such as kruppel like factor 4 (KLF4) and ATP binding cassette subfamily A member 1 (ABCA1) (44). Like DNA methylation, RNA is also methylated. The most abundant modification is the methylation of adenosine in mRNA (N6-methyladenosine, M6A), which is catalysed by the enzyme methyltransferase-like 3 (METTL3) (47). N6-methyladenosine is also found in other RNAs, such as transfer RNA, ribosomal RNAs, and other non-coding RNAs (48). Even though RNA methylation profile changes with age are not elucidated, it has been linked to cardiac hypertrophy, which is frequently observed with ageing (48). Dorn et al. demonstrated cardiac hypertrophy in METTL3 overexpressing mice by 8 months of age. The mechanism by which METTL3 overexpression causes hypertrophy is still being explored. However, initial findings indicate it increases the expression of the β-myosin heavy chain (β-MHC), a critical hypertrophic protein. In vitro analysis of METTL3 inhibition attenuates cardiomyocyte hypertrophy, indicating a role in the development of cardiac hypertrophy (48).

Histones are proteins which wrap DNA to form the nucleosome. The epigenetic modifications of histones, such as methylation, acetylation, carbonylation, and phosphorylation, are associated with CVDs (49). Like DNA methylation, histone modifications repress transcription (47). Histone methylation is mediated by histone methyltransferases and demethylases, which add and remove methyl groups from histones, respectively (47). Changes in the histone methylation profile have been observed in cardiac hypertrophy, atherosclerosis, and dilated cardiomyopathy. SET and MYND domain-containing protein 1 (SMYD1) is a histone methyltransferase which represses the activity of pro-hypertrophic genes such as transforming growth factor-β-3 (TGFβ3) and natriuretic peptide A (NPPA) and its depletion is associated with hypertrophy (49). However, there is no conclusive evidence that the histone methylation pattern changes with age. Histone acetyltransferases add an acetyl group from acetyl-CoA to lysine residues, and deacetylases remove the acetyl group (47). Histone deacetylase (HDAC) is catalysed primarily by HDAC1 and HDAC2, and their global deletion has been linked to cardiac arrhythmia and dilated cardiomyopathy due to the abnormal growth of cardiomyocytes (50). Whether changes in histone acetylation lead to age-associated CVDs, is not fully understood and is an area of active research (47).

Mitochondria have a double membrane consisting of inner and outer layers. The inner layer comprises 4 complexes—Complex I to IV, which generate ATP via oxidative phosphorylation (51). With age, oxidative phosphorylation declines primarily because of the dysfunction of complexes I and IV, reducing the production of ATP in the cells (52). This dysfunction can be due to the accumulation of damage to the mitochondrial DNA (mtDNA) and excess amounts of reactive oxygen species (ROS) (53). Increased ROS accelerates atherogenesis by oxidising low-density lipoprotein, promoting endothelial cell senescence and apoptosis (54). Oxidised low-density lipoproteins increase the risk of plaque formation and subsequent atherosclerosis by reducing the availability of nitric oxide and promoting inflammation, leukocyte adhesion, and smooth muscle cell proliferation (55). Increased ROS-linked endothelial cell apoptosis also impairs blood flow to the heart and the brain due to rarefaction of the blood vessels (56). ROS is also associated with cardiomyocyte necrosis and the progression of cardiac hypertrophy (57, 58). ROS generated by monoamine oxidase (MAO), an enzyme present on the outer membrane of the mitochondria, induces cardiomyocyte necrosis by activating p53 (10, 58). ROS also induces cardiac hypertrophy through the activation of pro-hypertrophic signalling kinases and transcription factors such as mitogen-activated protein kinase (MAP-kinase) and nuclear factor kappa light chain enhancer of activated B cells (NF-κB) (59, 60). Thus, increased ROS generation due to age-dependent mitochondria dysfunction is a crucial molecular change associated with cardiac ageing and the development of CVDs (59).

Mitochondria is a unique cell organelle, as it has its DNA known as mitochondrial DNA (mtDNA). mtDNA is circular and has genes that encode proteins necessary to produce energy (61). The mtDNA accumulates mutations with age, which can also be caused due to age-impaired production of excess ROS (62). mtDNA mutations have been linked with several diseases, such as Alzheimer's disease, Parkinson's disease, CVDs, diabetes, and cancer (63). Lindroos et al. studied the effect of one of the mtDNA mutations (m.3243A > G) in 14 individuals. This mutation affects the translation of respiratory chain complex proteins by impairing the tRNA necessary for the translation (64, 65). The study revealed impaired glucose oxidation and lower stroke volume in individuals with mutation (65). Studies have also identified an association of other mtDNA mutations, such as m.3256C > T, m.12315G > A, m.13513G > A, and m.15059G > A with atherosclerosis due to impaired translation of proteins essential for oxidative phosphorylation and increased production of ROS (66). While studies suggest a strong correlation between mtDNA, ageing, and CVD, whether mtDNA is the cause of CVD remains unclear, warranting future studies to ascertain the mechanisms and its reliability as an age-associated marker of CVDs (63).

Ageing has a detrimental effect on the neuro-hormonal signalling pathways essential for the functioning of the cardiovascular system. One of the signalling pathways linked to the age-dependent progression of cardiac function decline and CVD is the renin-angiotensin-aldosterone system (RAAS) (59). RAAS regulates blood pressure by controlling fluid and electrolytes through a well-coordinated balance between the liver, kidneys, heart, lungs and blood vessels (67). In brief, RAAS is activated in the kidneys by a drop in blood pressure, leading to renin's activation. Active renin is secreted in the blood, which converts inactive angiotensinogen (produced by the liver) to angiotensin I. Angiotensin I is converted by an angiotensin-converting enzyme (primarily present in the lungs) to angiotensin II (Ang II). Ang II is the central effector molecule of the RAAS system, which acts on the adrenal gland, brain, and blood vessels to elevate blood pressure by various mechanisms (68). Ageing increases Ang II concentration, which is associated with cardiac hypertrophy, fibrosis, and increased ROS (68). Another significant signalling pathway affected by age is the insulin-like growth factor-1 (IGF), which reduces with age in humans. While the reduction in the IGF signalling pathways has been observed to improve cardiac function with age in Drosophila and mice models, the same has not been observed in humans (59). Vasan et al. examined the levels of IGF-1 in 717 elderly individuals with no known history of myocardial infarction and congestive heart failure for a mean period of 5.2 years. Of the 717 elderlies, 56 developed congestive heart failure (69).

Interestingly, all these individuals had serum IGF-1 levels below the median value—the exact reason for these contradictory findings between different species of ageing warrants further studies (59). Beta-adrenergic signalling is an important signalling pathway which regulates heart rate and contractility. There are 3 major beta-adrenergic receptors—β1, β2, and β3 receptors, which are G-coupled protein receptors. In brief, ligands (major catecholamines such as epinephrine and norepinephrine) bind to the beta receptors, which activate adenylyl cyclase and increase the level of cyclic adenosine monophosphate (cAMP). cAMP targets protein kinase A, which activates several proteins such as ryanodine receptors, L-type calcium channels and myosin binding protein-C, eventually leading to cardiomyocyte contraction (70). With age, the beta-adrenergic signalling reduces, primarily due to impaired agonist binding to the receptor and reduced β1 receptors (70, 71). A decreased beta-adrenergic response seems beneficial, especially in the aged heart, as it reduces the risk of arrhythmia, hypertrophy, and apoptosis (70, 72). However, the lower beta-adrenergic response is also associated with lowered exercise tolerance, impaired autonomic regulation and impaired arterial-ventricular load (70, 72, 73).

Cellular senescence is a complex, dynamic, and multi-step process in which a cell undergoes permanent cell cycle arrest with continued metabolic activities (74). Cellular senescence is induced as a response to stressors such as DNA damage, telomere shortening, oxidative stress, mitochondrial damage, which are a result of ageing (74, 75). A senescent cell has multiple characteristic features such as an enlarged and flattened morphology, senescence associated beta-galactosidase activity, activation of senescence associated secretory phenotype (SASP), and DNA damage response (74, 76). SASP is a prominent hallmark of a senescent cell, as it releases pro-inflammatory chemokines and cytokines, growth factors, ECM proteins, and activation of p16 and p21, which affects the neighboring cells or distant cells, if they are released in the systemic circulation (76). SASP has a deleterious effect, as it can induce senescence in the neighboring healthy cells and even 10%–15% of senescent cells in a tissue, can cause tissue degeneration (77, 78). Cellular senescence has been linked to age-related diseases such as osteoporosis, renal diseases, neurodegenerative diseases, pulmonary fibrosis, and CVDs (79).

The heart is a mosaic of different cell types such as cardiomyocytes, vascular smooth muscle cells, endothelial cells, and fibroblasts, and senescence of individual cell types can cascade to a disease (80). For example, senescent cardiomyocytes along with the phenotypical changes of a senescent cell exhibit contractile dysfunction and hypertrophic growth, which eventually leads to cardiac hypertrophy, arrhythmias, cardiac remodeling, and heart failure (81). Senescent endothelial cells have impaired production of endothelin-1 (vasoconstrictor) and nitric oxide (vasodilator) which affects vascular function. Senescent endothelial cells have been studied to develop disorders such as atherosclerosis and heart failure with preserved ejection fraction (80). Senescent fibroblasts secrete IGF-1 which promotes collagen synthesis and exacerbates cardiac fibrosis. It also induces senescence in neighboring cardiomyocytes through paracrine signalling (80). Similarly, senescent vascular smooth muscles lead to atherosclerosis and pulmonary hypertension (82).

Senolytics are a class of drugs which selectively induces apoptosis in senescent cells by activating the B-cell lymphoma 2 (BCL-2) family proteins, p53, phosphoinositide-3-kinases (PI3K) and other apoptotic pathways (83). Salerno et al. combined Dasatinib and Quercetin, two senolytic drugs and administered to 22–24 months old mice after acute MI, released healthy cardiac stem cells and improved cardiac remodeling and regeneration, which eventually improved LV function. Thus, removal of senescent cells by senolytics improves the overall functioning of the heart, even in aged mice (77, 83). In a recent study Cattaneo et al. reported the importance of longevity-associated BPIFB4 (LAV-BPIFB4) gene in supporting cardiac function and vascularization in ageing cardiomyopathy (84). They demonstrated reduced LAV-BPIFB4 in older hearts and that gene therapy with LAV-BPIFB4 rescued cardiac function and myocardial perfusion in aged mice by improving microvasculature density and pericyte coverage. Therefore, therapeutic modalities targeting cellular senecesnce may have therapeutic potential although long-term studies required to determine the sustainability of the effect.

The human genome consists of only 3% protein-coding genes, whereas most of the genome is transcribed to produce non-coding RNAs (nc-RNAs) (85). These nc-RNAs include a wide array of molecules such as microRNAs (miRNAs), long-noncoding RNAs (lncRNAs), small nuclear RNAs, small nucleolar RNAs, piwi-interacting RNA, circular RNA and transfer RNA etc (86).. Based on their function, these nc-RNAs can be classified into two categories: housekeeping RNAs and regulatory RNAs. Housekeeping RNAs, such as ribosomal RNA, transfer RNA, and small nuclear RNA, are present in virtually all cells and are essential for the cell's normal functioning. Regulatory RNAs such as piwi-RNA, miRNAs, and lncRNAs regulate gene expression at the transcriptional or translational levels (86, 87). Among the ncRNAs, miRNAs and lncRNAs are gaining prominence in ageing-associated diseases. The expression of miRNAs and lncRNAs dysregulate with age which leads to the onset and progression of ageing-associated diseases such as CVDs (88, 89), Alzheimer's (90), cancer (91), and diabetes (92). Furthermore, miRNAs such as miR-146a/b, -126, -34a, -22 are associated with inducing senescence in vascular smooth muscle cells, endothelial cells, cardiomyocytes, and fibroblasts, respectively (80, 93). The role of lncRNAs with respect to senescence is an active area of research, although to date only few lncRNAs that induce cellular senescence have been identified. For example, senescence associated lncRNA-1 triggers senescence in fibroblasts and lncRNA-H19 triggers senescence in cardiomyocytes (89, 94). Several studies have demonstrated the critical role of miRNAs and lncRNAs in cardiovascular system homeostasis and that their dysregulation with age leads to CVDs (95, 96). In the next section, we will focus on non-coding RNAs (miRNAs and lnc-RNAs) in the age-dependent development of CVDs.

The biogenesis of miRNAs begins in the nucleus by the most common canonical or less common non-canonical pathway (summarised in Figure 2). In the canonical pathway, the primary transcript (pri-miRNA) transcribed by RNA-polymerase II is a several hundred nucleotides long transcript, which is processed to form a 70–100 nucleotide long precursor known as pre-miRNA (103, 104). Next, pri-miRNA is cleaved by a protein complex called the microprocessor, which comprises two enzymes—Drosha and DiGeorge syndrome critical region gene 8 (DGCR8) to form a shorter pre-miRNA. DGCR8, an RNA binding protein, binds to the pri-miRNA and Drosha cleaves it at the base of the hairpin structure, which results in a 2 nt 3′ overhang. Next, a nuclear transport receptor protein called exportin-5, along with RAN-GTPase, moves the pre-miRNA from the nucleus to the cytoplasm. Once in the cytoplasm, Dicer, an RNase III endonuclease, removes the terminal loop from the pre-miRNA to form the mature miRNA consisting of 18–22 nucleotides (104, 105). Finally, the mature miRNA duplex is loaded to RNA-induced silencing complex (RISC), a multi-protein complex consisting of Argonaute protein (Ago2). Once loaded on the RISC complex, one of the duplex strands, known as the passenger strand, is released, while the other strand, the guide strand, remains attached. The passenger strand can also be cleaved by Ago2 or C3PO endonuclease. The guide strand selection is based on the relative thermodynamic stability of the two strands. The guide strand generally has a relatively unstable 5′ end and has a uridine nucleotide in the first position (98). The loaded miRNA on the RISC (miR-RISC) binds to the complementary sequence of its target messenger RNA (mRNA). Partial complementarity between the miR-RISC and mRNA leads to translational repression, while an exact complementarity leads to mRNA cleavage and degradation (98).

In the non-canonical pathway of miRNA biogenesis, miRNAs can be formed either by the Drosha/DGCR8 independent pathway or the Dicer independent pathway. Mirtrons are an example of the miRNAs formed by the Drosha/DGCR8 independent pathway. The spliceosome cleaves the pri-miRNA of mirtrons, which forms a branched pre-miRNA. Next, the 3′ and 5′ ends of the pre-miRNA are ligated, forming a lariat, which is linearised by a debranching enzyme. After the debranching, the pre-miRNA can be transported to the cytoplasm by the exportin-5-RAN-GTPase complex. Once in the cytoplasm, it is cleaved by Dicer and loaded on Ago to form a mature miRNA (106, 107).

In the Dicer-independent pathway, the microprocessor complex produces the pre-miRNA as short-hairpin RNA transcripts, which are then exported to the cytoplasm. However, the exported transcripts are too short to be processed by Dicer; instead, the entire pre-miRNA duplex is loaded on the Ago complex and undergoes splicing by Ago2 to form the mature miRNA [104] (Figure 2).

While the role of miRNAs in CVD is very well established, it is only recently that researchers have started to explore the role of miRNA dysregulation in the development of age-induced CVD. Therefore, this section will review some key literature on cardiovascular-enriched microRNAs and their role in cardiac ageing (summarised in Figure 3).

The biogenesis of lncRNA is different from that of miRNAs but is similar to the formation of mRNAs. LncRNAs are transcribed by RNA polymerase II (89). Most of the lncRNA has a polyadenylated 3′ end and is capped with methyl-guanosine at the 5′ end (152). They undergo alternative splicing and modifications to form lncRNA (153). Based on the cellular fate and function of the specific lncRNA, it can either be localised in the nucleus or transported to the cytoplasm. LncRNAs are exported to the cytoplasm by the nuclear RNA export factor 1 (NXF1) (154). Once in the cytoplasm, according to their function, they are associated with ribosomes located in the mitochondria or exosomes. In general, the biogenesis of lncRNA varies significantly based on the function of the lncRNA (154).