Lakhasly

Genetic factors in aging The last two decades have been a revolution for human genetics, starting with the sequencing of a human genome in 2003 and breakthroughs in genome-wide association studies finding thousands of genetic loci associated with complex human traits, including many age-related diseases.It should also be noted that sex differences in dementia incidence may be partially explained by selective survival (Shaw et al., 2021).A recent analysis in UK Biobank found that abundant mtDNA, estimated from the weighted intensities of probes mapped to the mitochondrial genome, was significantly elevated in premenopausal women compared to men and inversely associated with age, smoking, BMI, and frailty (Hagg et al., 2020).Experiments in worms, flies, and mice have shown that overexpressing chaperones and heat-shock proteins are associated with an extended lifespan, whereas models deficient in parts of the chaperone-heat-shock system present accelerated aging phenotypes (Lopez-Otin et al., 2013; Ferrucci et al., 2020).The first views frailty as a physical syndrome with a discrete categorization of individuals into nonfrail, prefrail, and frail, whereas the latter considers frailty as a multidimensional construct based on the accumulation of deficits in physical, biological, and psychosocial domains.The key pathways include the insulin/insulin-like growth factor 1 (IGF-1) signaling pathway, mechanistic target of rapamycin (mTOR), and adenosine monophosphate-activated protein kinase (AMPK) pathway (Pignatti et al., 2020).Over the past years, there has been intensive research on how nutrient-sensing pathways control lifespan and healthspan, with the most significant breakthroughs achieved in unraveling how different dietary restrictions improve aging outcomes and survival in several species, including humans (Templeman and Murphy, 2018).For aging, gene discoveries have been sparse, although lately, large cohorts such as the UK Biobank have enabled powerful analyses of parental lifespan, healthspan, and longevity (Timmers et al., 2019; Zenin et al., 2019; Melzer et al., 2020).Recent findings in this area point toward sexual dimorphism in transcriptomic profiles with hormone-related regulation and associations with various processes such as tissue morphogenesis, fat metabolism, cancer, and immune responses (Oliva et al., 2020).For example, such sex-specific targeting patterns of transcription factors have been found for genes associated to Alzheimer's disease (AD), Parkinson's disease (PD), diabetes, autoimmune thyroid disease, and cardiomyopathy (Lopes-Ramos et al., 2020).Mitochondria-linked mechanisms Mitochondrial DNA (mtDNA) is inherited from mothers and contains the genetic code for 13 proteins, essential components of oxidative phosphorylation complexes, and several RNAs (Kauppila et al., 2017).The most well-studied mechanisms are acetylation and methylation processes; changes in histone acetylation/methylation have been linked to aging, healthspan, and lifespan in diverse models, such as flies, mice, yeast, and human cell lines (Yi and Kim, 2020).The former refers to changes in the adaptive immune system, such as increased numbers of memory CD8 +T cells (resulting in a decreased CD4/CD8 cell ratio), loss of the key costimulatory molecule CD28 on the T cell surface and compromised clonal expansion and specific antibody production in the B cell compartment (Gubbels Bupp, 2015; Franceschi, 2019).Between puberty and menopause - when differences in the hormonal milieu are the greatest between men and women - women experience lower rates of infections, an advantage attributed to stronger immune and vaccine responses and more efficient pathogen clearance (Gubbels Bupp, 2015).These pathways regulate a multitude of intracellular functions, such as cell cycle control, DNA replication and repair, autophagy, and antioxidant defenses, by which their effects are excreted for reproduction, growth, and aging (Pignatti et al., 2020).Events such as critically short telomeres, oxidative stress, replicative errors, mitochondrial dysfunction, pathogen response, oncogene activation, and other stress sources may induce senescence of the cell with irreversible replicative arrest (Lopez-Otin et al., 2013).During aging, the balance in the protein machinery is lost and unfolded and misfolded proteins can aggregate and cause pathological conditions seen in diseases of (neuro)degeneration, AD, PD, and diabetes (Hipp et al., 2019).Analyses on sexual dimorphism in human protein homeostasis and autophagy aging processes are still lacking, which is understandable, as efficient high-throughput methods are not yet available (Ferrucci et al., 2020; Pomatto et al., 2017).Studies investigating genome-wide DNA methylation differences between men and women report significant differences in autosomes and on the X chromosome, the latter being linked to sexual dimorphism genes and XCI (Li et al., 2020c; McCartney et al., 2019).Inflammaging refers to chronic, low-grade inflammation that occurs in the absence of infection and manifests as increased production of proinflammatory cytokines, linked to both frailty and CVD (Ferrucci and Fabbri, 2018).Sex differences are present in diverse species ranging from insects to mammals, with female individuals presenting stronger innate and adaptive immune responses than males (Klein and Flanagan, 2016).Each of the hallmarks of aging is associated with undesirable metabolic alterations (Lopez-Otin et al., 2016), stressing the fact that nutrient sensing and metabolism are interlinked processes with broad effects on whole-organism functions.Interestingly, genetic polymorphisms in genes encoding proteins in the insulin/IGF and mTOR pathways are among those that are robustly associated with longevity, such that variants associated with the lower basal activity of the pathways are associated with longevity (Pan and Finkel, 2017).One of the most commonly used and strongest markers for human population-based estimation of death risk is a simple assessment of walking speed (Ganna and Ingelsson, 2015), yet other popular measures include grip strength, chair rise, lung function, vision, and an abundance of cognitive domains (Peiffer et al., 2010).Sex-specific differences are seen across almost all respiratory structures and functions; women have smaller and anatomically different lungs than men, perform worse in breathing exercises, and sex hormones interact with lung and airway function during early developmental processes and aging (LoMauro and Aliverti, 2018).Perhaps, the sex specificity in functional measures best describes the complex interplay between fitness and aging in line with the rate of living theory in the senescence theory of aging, emphasizing the sex paradox in aging where women with worse physical function and health still outlive men, possibly due to a better cellular maintenance system and protections from estrogens.However, CVD is also tightly linked to inflammaging of the vasculature, and cellular senescence that could be reflected as TL shortening and intrinsic epigenetic age acceleration (Ferrucci and Fabbri, 2018).Few efforts have been made for X chromosome-wide association studies, but they reveal (sex-specific) links to several complex traits and identify a locus associated with height escaping XCI (Bernabeu, 2020; Tukiainen et al., 2014).The X chromosome encodes approximately a thousand genes, many related to metabolic activity, such as amino acid turnover and transport, and could explain the differential proliferative rates in sexes seen during embryonic growth (Patrat et al., 2020).This state causes a response of cytokines and other proinflammatory factors to be released, which may trigger downstream effects in the surrounding tissue and invoke a senescence-associated secretory phenotype (SASP) (Ferrucci et al., 2020).Hence, cellular senescence is a core mechanism in the senescence theory of aging, where cells and tissues accumulate damage across life but is also essential in the Hayflick limit's programmed theory of aging (Ferrucci et al., 2020; Khosla et al., 2020; Schafer et al., 2020).Genetic studies of epigenetic clocks have discovered several loci associated with lifespan and lifestyle factors beyond the gene regions where the CpGs themselves are located (Lu et al., 2018; McCartney, 2020).Moreover, intrinsic and extrinsic epigenetic clocks have been suggested to represent internal (cellular) versus external (lifestyle stressor) aging processes (Horvath and Raj, 2018).From an evolutionary perspective, inflammaging can result from positive selection of genetic variants that associate with higher levels of pro-inflammatory factors and enhanced immune responses in early life, conferring better protection against pathogens but resulting in increased damage to host tissues in later life.Inflammaging is thus in accordance with multiple different theories, where various stimuli, such as oxidative stress and lifestyle factors, contribute as well (Franceschi, 2019; De la Fuente and Miquel, 2009).As stated above, men seem to experience immunosenescence to a greater extent than women, potentially because women exhibit higher basal immunoglobulin levels, higher CD4 +T cell counts, and an increased CD4/CD8 T cell ratio compared to men (Gubbels Bupp, 2015; Gomez et al., 2018).The corresponding adaptive immune functions, such as antigen-specific antibody responses and CD4 +T cell cytokine production, are also typically more enhanced in women (Gubbels Bupp, 2015; Gomez et al., 2018).Of the different dietary restrictions, the most compelling evidence rests on caloric restriction (CR), in which the energy intake is reduced ~30% relative to ad libitum-fed animals without reducing the intake of micronutrients (Templeman and Murphy, 2018).Sex hormones regulate several key functions in nutrient sensing and metabolism of glucose, amino acids, and proteins, and it is not surprising that men and women differ in several metabolic characteristics.Akin to epigenetic clocks (see Epigenetic alterations) that predict mortality independent of other risk factors, there have been attempts to create similar composite measures based on metabolites measured using different techniques (Jylhava et al., 2017).Hence, little is known about sex-specific genetic effects for complex traits, although sexual dimorphisms have been reported for anthropometric traits and gout (Bernabeu, 2020; Randall et al., 2013), and gene-sex interactions have been found for multiple sclerosis (Traglia et al., 2017).During aging, the XCI ratio between maternal and paternal X chromosomes is no longer equal, leading to skewed XCI, which has been implicated in diseases and shown to be less severe in female centenarians (Gentilini et al., 2012).Studies in rodents and in Drosophila support the association between DNA repair, mutational burden, and aging; however, sex-specific effects are intricate, and the results depend heavily on animal strain and environmental conditions (Fischer and Riddle, 2018).Taken together, the examples described here relate to chromosomal stability and resemble well with the senescence theory of aging, where random events occur over time with less capacity of our maintenance system to repair and fix the faults.Sexual dimorphism in genome-wide studies for anthropometric traits may be consistent with the developmental processes and growth controlled during early life and aging, where hormonal influences are also apparent.Oxidative damage and increased ROS production across life were initially thought to cause this dysfunction, but research in recent years showed that ROS do not accelerate aging in mice and even prolong lifespan in yeast and C. elegans (Lopez-Otin et al., 2013).In the elderly, cellular senescence is apparent where DNA maintenance and repair are no longer efficient, and telomeres reach critical lengths for cellular survival consistent with a person's natural lifespan limit (Steenstrup et al., 2017).Recently, a model using hyperlong telomeres showed that this phenotype increases the lifespan in mice and shows overall beneficial effects on metabolism, glucose control, and mitochondrial function (Munoz-Lorente et al., 2019).A mutation in the DKC1 gene on the X chromosome - a gene involved in telomere biology - is often seen in patients with dyskeratosis congenita, which leads to rapid TL shortening and reduced survival (Savage and Bertuch, 2010).However, a recent meta-analysis investigated sex differences in TL across 51 vertebrate species and found no evidence supporting either the heterogametic sex disadvantage or the sexual selection hypotheses (Remot et al., 2020).Cellular senescence is tightly linked with aging, has been well correlated with DNA damage, and an increasing number of cells are senescent in old tissues compared to young tissues in a study of liver tissue in mice (Lopez-Otin et al., 2013; Khosla et al., 2020).Nevertheless, the systemic accumulation of senescent cells in aging has been associated with many age-related diseases and conditions, such as frailty, both in humans and animal models (Ferrucci et al., 2020; Khosla et al., 2020; Schafer et al., 2020).A study by Klein et al., 2019 found that tau may affect histone acetylation in the human brain using an epigenome-wide association study of H3K9ac, thus relating histone modification processes to AD pathology.While both sexes experience aging-associated changes in the immune system, the hallmark features differ for men and women, and men are considered to experience maladaptive changes to a greater extent (Gubbels Bupp, 2015; Gomez et al., 2018).The FAI includes muscle strength (grip strength), movement (gait speed), sensory (vision and hearing), and lung function and is predictive of mortality in both sexes, yet the hazard ratio is greater in women (Finkel et al., 2019).Although frailty often coexists with multimorbidity (and disability), the association between frailty and mortality is independent of multimorbidity (Hanlon et al., 2018), indicating that frailty captures health-related variation that is not attributed to diseases alone.There is currently no widely accepted consensus on how to measure frailty; however, the two most commonly used approaches are the Fried phenotypic model (FP) (Fried et al., 2001) and the Rockwood frailty index (FI) (Searle et al., 2008).However, another theory suggested underlying the sex differences is the chronic disease hypothesis by which women are more likely to experience nonlethal chronic conditions, while men tend to develop acute conditions associated with high mortality, such as stroke and myocardial infarction (Gladyshev, 2014; Bernabeu, 2020).Hence, genomic instability, including the accumulation of mutations, epigenetic alterations, and telomere attrition, are hallmarks of aging and provide a link between aging and sexual dimorphism mechanisms in cancer.Cognitive aging in healthy adults demonstrates sex-differential effects, where men generally perform better in visuospatial ability and women better in verbal ability, but the speed of decline may be worse in men, although the literature is not consistent (Li et al., 2020b; McCarrey et al., 2016).Hence, many genes may be linked to the underlying aging process or beneficial for age-related diseases, with importance for longevity, health, and lifespan, but they may not have been specifically selected for (Rattan, 2000).Rare somatic mutations may accumulate across life and play a role in cancer, where men have been shown to have mutation accumulation earlier in life (Podolskiy et al., 2016), and in several premature-aging syndromes (Fischer and Riddle, 2018).In humans, studies have linked the accumulated burden of mutations in mtDNA to aging and PD, although a majority of the mtDNA molecules within a cell must be affected for critical symptoms to emerge (Kauppila et al., 2017).In humans, women show higher mitochondrial gene expression levels, protein content, and overall activity in multiple tissues, such as the brain, skeletal muscle, and cardiomyocytes (Ventura-Clapier et al., 2017).There are four major types of epigenetic mechanisms: ATP-dependent chromatin remodeling complexes, histone, and DNA modifications, and noncoding RNAs (Pagiatakis et al., 2021).Moreover, in women, earlier menopause, either natural or surgical, is associated with increased epigenetic age, and although the finding was not consistent across different tissues, there was further support for lower epigenetic age in women undergoing HRT (Levine et al., 2016).However, after the age of menopause, the incidence of autoimmune diseases in women decreases close to the numbers observed in men, whereas the incidence of chronic inflammatory diseases increases (Gubbels Bupp, 2015).A recent study using sequencing and flow cytometry data in blood mononuclear cells further elucidated the sexual dimorphism in immune aging by showing that male and female cells also significantly differ at the age when sex hormones decline (Marquez et al., 2020).Frailty Frailty is defined as a state of increased vulnerability to stressors resulting from decreased physiological reserves to maintain homeostasis across multiple organ systems.(For a more in-depth discussion on sex and gender aspects in aging diseases and treatment, we refer the reader to Mauvais-Jarvis et al., 2020 and Regitz-Zagrosek, 2012).Longevity is known to be moderately heritable (Melzer et al., 2020); however, from an evolutionary perspective, natural selection is active for the reproduction of a species and not for maximizing lifespan.Another recent study found measured and genetically predicted levels of ten serum biomarkers to be associated with healthspan and lifespan, and again with stronger effects for healthspan seen in women (Li et al., 2021).However, evidence from both human and animal studies points toward the fact that the accumulation of mtDNA mutations is a feature of early life replication errors that undergo polyclonal expansion independent of ROS (Lopez-Otin et al., 2013).Substantial sexual dimorphism has been observed for mitochondrial function concerning oxidative capacity and enzyme activity (Ventura-Clapier et al., 2017).Therefore, throughout life, the length of the telomere (TL) decreases and serves as a marker of cellular aging (Blackburn et al., 2015).As such, short TL has been associated with age-related outcomes and health aspects, for example mortality (Wang et al., 2018), CVD (Haycock et al., 2014), and different stressors in life (Starkweather et al., 2014).Different genetic models have been used in mice to lengthen telomeres with telomerase activation, where some experiments increased the cancer incidence, while others did not (Folgueras et al., 2018).These findings have led to increased knowledge, and many studies have provided evidence for causal associations between short leukocyte TLs and age-related diseases (Kuo et al., 2019).Genetic liability contributes to the overall length of telomeres in all cells, and across the lifespan, stressors and lifestyle factors influence cell-specific attrition rates.Evidence points to the fact that female stem cells have an increased capacity for regeneration, self-renewal, and proliferation (Dulken and Brunet, 2015), in line with a more beneficial cellular aging route in females/women.Proteostasis and autophagy Protein homeostasis, or proteostasis, is the body's ability to maintain control over protein synthesis, folding, stability, degradation, and removal through autophagy (Hipp et al., 2019).A study by Ubaida-Mohien et al. found a decreased representation of chaperone proteins in old skeletal muscle tissue in healthy adults, although autophagy-related proteins were overrepresented (Ubaida-Mohien et al., 2019).In model systems, abrogation of autophagy leads to neurodegeneration and shortens lifespan, whereas increased basal activity of autophagy increases lifespan (Leidal et al., 2018).Hence, declining proteostasis control in aging may be an effect of accumulated aggregates and dysfunctional autophagy, consistent with the senescent wear-and-tear theory of aging, including the ROS theory of aging.However, in human studies, genome-wide DNA methylation arrays have paved the way for a new field of research on epigenetic age where hundreds of (un)methylated sites (CpGs) have been shown to associate with age across the life course (Zhang et al., 2020).A multitude of clocks quantifying biological age across tissues, in whole blood, skin, muscle, or in human cell culture models have emerged (Horvath and Raj, 2018) and recently across mammalian species (Lu, 2021).With remarkable accuracy, clock ticks with aging and a higher epigenetic age are associated with worse health and increased mortality risk (Horvath and Raj, 2018; Chen et al., 2016).Little is known about sex-dimorphic effects on histone modifications in aging, although studies on different interventions and acetylation/methylation in animals suggest that these effects are important modifiers in aging (Fischer and Riddle, 2018).The temporal dynamics of these changes point to the crucial role of sex hormones in shaping immune aging, although it is likely much more complicated, involving an interplay of multiple homeostatic systems.Like humans, the differences are largely attributable to the effects of sex hormones, with a contribution of genetic differences due to several immunoinflammatory genes that are X chromosome encoded (Klein and Flanagan, 2016).At the molecular level, women have lower fasting insulin and glucose levels, lower basal fat oxidation, and higher fat use but lower consumption of carbohydrates during physical activity (Comitato et al., 2015).Cardiac remodeling due to aging is universal, but the decline in myocytes and systolic function are greater in males, both in humans and rodents (Keller and Howlett, 2016).However, evidence supporting this conception is not conclusive (Merrill et al., 1997; Macintyre et al., 1999), and the underlying mechanisms for the sex-frailty paradox remain unresolved.Few studies have reported that aged female mice exhibit higher FI scores than males (Heinze-Milne et al., 2019).However, there is some evidence that while most transcription factors have similar expression profiles in men and women, there may be sex-specific regulatory networks across different tissues, leading to altered function and disease control (Lopes-Ramos et al., 2020).Telomeres Telomeres are repeated sequences of nucleotide bases (TTAGGG)n located at the end of the chromosomes (Blackburn et al., 2015).Although most telomere-related genes have been found in autosomal chromosomes, it has been suggested that the unguarded chromosome in heterogametic sex is a disadvantage for mortality and telomere maintenance.Currently, there is also increasing evidence for the beneficial antiaging effect of senolytic drugs as potential treatments to remove senescent cells when abundant (Ferrucci et al., 2020).In humans, long-lived families have a better-maintained autophagy system, and individuals under starvation exhibit enhanced autophagic flux (Leidal et al., 2018).Females studied in animals and model systems also exhibit more resistance to stressors, partly hypothesized to be due to estrogens' beneficial effects (Tower et al., 2020).As with telomeres, the epigenetic clocks seem to be tightly linked with cellular replication underlying the Hayflick limit theory (Wagner, 2019).Taken together, the sexual dimorphism seen in epigenetic studies on aging is complex and seems to reflect sex chromosome-linked mechanisms and/or hormonal biological processes.Although animal models cannot fully recapitulate human immunosenescence or inflammaging, findings on sex-related immune functions in animal studies have generally been in line with observations in humans.Nutrient sensing Intracellular nutrient-sensing pathways and signaling systems mediate information on nutrient availability and energy levels in the extracellular milieu.At the molecular level, CR triggers activation of stress response pathways that in turn reduce inflammation and increase repair and antioxidative functions.However, the higher basal insulin levels in men promote glycogen and lipid synthesis in muscle cells, resulting in higher muscle mass and strength (Comitato et al., 2015).Analogous to the Hayflick limit, the rate of aging theory posits that the total amount of energy expenditure per lifetime is finite and that excessive usage results in accelerated aging (Pearl, 2011).A recent study created a composite measure, termed the functional aging index (FAI), to better capture the state and changes in various physical functions simultaneously.As men have higher muscle mass than women, some clues might be obtained from the observed associations between higher skeletal muscle mass and higher basal metabolic rate, that is energy expenditure that is higher in men than in women (Ruggiero et al., 2008).In recent years, animal models of frailty, building on both FI and FP, have become available, providing opportunities to untangle how and why frailty develops and the mechanisms behind the sex differences.However, other studies have reported no difference between the sexes, and one study found that male mice had higher FI scores than females (Heinze-Milne et al., 2019).The top 10 leading causes of death by sex in those above 70 years reveals a change in the ranking of diseases so that instead of colon and rectum cancers, prostate cancer emerges in men and communicable diarrheal diseases in women.Similar sexual dimorphism is observed in rodent models investigating mitochondrial respiratory function (Ventura-Clapier et al., 2017).Figure 2

Genetic factors in aging
The last two decades have been a revolution for human genetics, starting with the sequencing of a human genome in 2003 and breakthroughs in genome-wide association studies finding thousands of genetic loci associated with complex human traits, including many age-related diseases. For aging, gene discoveries have been sparse, although lately, large cohorts such as the UK Biobank have enabled powerful analyses of parental lifespan, healthspan, and longevity (Timmers et al., 2019; Zenin et al., 2019; Melzer et al., 2020). However, only a handful of genes have been identified, and the top loci are often well known for their relation to diseases, for example APOE, LPA, and CDKN2B-AS1. Longevity is known to be moderately heritable (Melzer et al., 2020); however, from an evolutionary perspective, natural selection is active for the reproduction of a species and not for maximizing lifespan. A recent study using human genotype data found that rare germline mutational burden was associated with lifespan and healthspan (Shindyapina et al., 2020). In particular, the association between mutations and healthspan was more pronounced in women. Another recent study found measured and genetically predicted levels of ten serum biomarkers to be associated with healthspan and lifespan, and again with stronger effects for healthspan seen in women (Li et al., 2021). Hence, many genes may be linked to the underlying aging process or beneficial for age-related diseases, with importance for longevity, health, and lifespan, but they may not have been specifically selected for (Rattan, 2000).

Thus far, large-scale genome-wide association studies have focused mostly on autosomes and rarely even stratified results by sex. Hence, little is known about sex-specific genetic effects for complex traits, although sexual dimorphisms have been reported for anthropometric traits and gout (Bernabeu, 2020; Randall et al., 2013), and gene-sex interactions have been found for multiple sclerosis (Traglia et al., 2017). Few efforts have been made for X chromosome-wide association studies, but they reveal (sex-specific) links to several complex traits and identify a locus associated with height escaping XCI (Bernabeu, 2020; Tukiainen et al., 2014). The mechanisms by which XCI is controlled are complex, for example by noncoding RNA and epigenetics (Lee, 2011), and are a way to balance the unequal amount of X-chromosomal DNA between men and women. The X chromosome encodes approximately a thousand genes, many related to metabolic activity, such as amino acid turnover and transport, and could explain the differential proliferative rates in sexes seen during embryonic growth (Patrat et al., 2020). During aging, the XCI ratio between maternal and paternal X chromosomes is no longer equal, leading to skewed XCI, which has been implicated in diseases and shown to be less severe in female centenarians (Gentilini et al., 2012). For men, the mosaic LOY in blood cells increases with age and is associated with age-related diseases and a higher risk of death (Forsberg, 2017).

Although the sex chromosomes are responsible for most of the female and male-specific traits, autosomes have bene increasingly studied for their role in sex-specific gene expression and associations with biological functions. Recent findings in this area point toward sexual dimorphism in transcriptomic profiles with hormone-related regulation and associations with various processes such as tissue morphogenesis, fat metabolism, cancer, and immune responses (Oliva et al., 2020). Implications on immunoinflammatory functions have also been highlighted (Bongen et al., 2019; Nevalainen et al., 2015). The underlying mechanisms for the sex differences in tissue specific transcription and associations with disease risks in men and women is currently unclear. However, there is some evidence that while most transcription factors have similar expression profiles in men and women, there may be sex-specific regulatory networks across different tissues, leading to altered function and disease control (Lopes-Ramos et al., 2020). For example, such sex-specific targeting patterns of transcription factors have been found for genes associated to Alzheimer’s disease (AD), Parkinson’s disease (PD), diabetes, autoimmune thyroid disease, and cardiomyopathy (Lopes-Ramos et al., 2020).

Genomic instability, such as chromosomal abnormalities, is known to be one of the hallmarks of biological aging (López-Otín et al., 2013). DNA damage accumulates across the life course as exogenous and endogenous triggers occur and DNA repair mechanisms become less efficient. Rare somatic mutations may accumulate across life and play a role in cancer, where men have been shown to have mutation accumulation earlier in life (Podolskiy et al., 2016), and in several premature-aging syndromes (Fischer and Riddle, 2018). Studies in rodents and in Drosophila support the association between DNA repair, mutational burden, and aging; however, sex-specific effects are intricate, and the results depend heavily on animal strain and environmental conditions (Fischer and Riddle, 2018). Taken together, the examples described here relate to chromosomal stability and resemble well with the senescence theory of aging, where random events occur over time with less capacity of our maintenance system to repair and fix the faults. Sexual dimorphism in genome-wide studies for anthropometric traits may be consistent with the developmental processes and growth controlled during early life and aging, where hormonal influences are also apparent.

Mitochondria-linked mechanisms
Mitochondrial DNA (mtDNA) is inherited from mothers and contains the genetic code for 13 proteins, essential components of oxidative phosphorylation complexes, and several RNAs (Kauppila et al., 2017). Mitochondria are important for cellular processes such as energy production, oxidation, and apoptosis, and their function has been described as one of the hallmarks of aging (López-Otín et al., 2013). Mitochondrial dysfunction is associated with many age-related diseases (Ferrucci et al., 2020; Chocron et al., 1865). Oxidative damage and increased ROS production across life were initially thought to cause this dysfunction, but research in recent years showed that ROS do not accelerate aging in mice and even prolong lifespan in yeast and C. elegans (López-Otín et al., 2013). In humans, studies have linked the accumulated burden of mutations in mtDNA to aging and PD, although a majority of the mtDNA molecules within a cell must be affected for critical symptoms to emerge (Kauppila et al., 2017). Another feature of aging is the number of mtDNA copies within a cell. A lower number has been associated with aging, cognitive and physical decline, and increased mortality (Mengel-From et al., 2014). Historically, the free radical theory of aging, or the ROS theory of aging, has been postulated to explain mitochondrial dysfunction in aging (Gladyshev, 2014). However, evidence from both human and animal studies points toward the fact that the accumulation of mtDNA mutations is a feature of early life replication errors that undergo polyclonal expansion independent of ROS (López-Otín et al., 2013). The latter fits well with the whole senescence theory of aging (in which energy needs to be preserved to last across the full lifespan) and the mutation accumulation theory.

Substantial sexual dimorphism has been observed for mitochondrial function concerning oxidative capacity and enzyme activity (Ventura-Clapier et al., 2017). In humans, women show higher mitochondrial gene expression levels, protein content, and overall activity in multiple tissues, such as the brain, skeletal muscle, and cardiomyocytes (Ventura-Clapier et al., 2017). Similar sexual dimorphism is observed in rodent models investigating mitochondrial respiratory function (Ventura-Clapier et al., 2017). Estrogens have been shown to influence mitochondrial function and exert protective effects, partly explaining why women have delayed mitochondrial aging compared to men. These differences may contribute to altered mitochondrial function during stress conditions such as injury or starvation, where sex-specific effects are also noted on mitochondrial respiration (Demarest and McCarthy, 2015). Little is known about sexual dimorphisms in mtDNA copy numbers and accumulated mutations in relation to aging. A recent analysis in UK Biobank found that abundant mtDNA, estimated from the weighted intensities of probes mapped to the mitochondrial genome, was significantly elevated in premenopausal women compared to men and inversely associated with age, smoking, BMI, and frailty (Hägg et al., 2020). Hence, taken together, sex hormones likely play a pivotal role in explaining the beneficial effect seen in women on mitochondrial function and aging.

Telomeres
Telomeres are repeated sequences of nucleotide bases (TTAGGG)n located at the end of the chromosomes (Blackburn et al., 2015). Every time a cell divides, the DNA polymerase machinery replicates the DNA sequence into two identical copies, although the last part of the DNA is not preserved due to the end replication problem. Hence, instead of losing important coding materials, the telomere is shortened. When it becomes critically short, the cell enters senescence, and this was later found to be the explanation for the Hayflick limit (Olovnikov, 1996). However, germline cells have an active telomerase enzyme that elongates the telomeres to maintain length, as do many cancer cells, but somatic cells do not normally have this process. Therefore, throughout life, the length of the telomere (TL) decreases and serves as a marker of cellular aging (Blackburn et al., 2015). As different cells have varied rates of cellular turnover, the attrition rates of telomeres depend on the proliferative capacity of the host cell. Leukocyte TL is among the most proliferative cells with fast TL shortening, while skeletal muscle maintains longer telomeres (Demanelis et al., 2020). Increased attrition rates are seen in childhood and adolescence, when growth and development occur, as well as in old adults. In the elderly, cellular senescence is apparent where DNA maintenance and repair are no longer efficient, and telomeres reach critical lengths for cellular survival consistent with a person's natural lifespan limit (Steenstrup et al., 2017). As such, short TL has been associated with age-related outcomes and health aspects, for example mortality (Wang et al., 2018), CVD (Haycock et al., 2014), and different stressors in life (Starkweather et al., 2014). Telomeres are present across many species, but their length and attrition rates may vary (Oeseburg et al., 2010). Different genetic models have been used in mice to lengthen telomeres with telomerase activation, where some experiments increased the cancer incidence, while others did not (Folgueras et al., 2018). Recently, a model using hyperlong telomeres showed that this phenotype increases the lifespan in mice and shows overall beneficial effects on metabolism, glucose control, and mitochondrial function (Muñoz-Lorente et al., 2019).

The lengths of the telomeres are also sex-specific. At birth, boys have shorter TLs than girls (Factor-Litvak et al., 2016), which prevails throughout life (Gardner et al., 2014). As women have a longer lifespan than men, telomeres have been suggested as the causal factor explaining the difference. However, it is still not completely understood whether telomeres could be the cause or consequence of biological processes. Several large-scale genomic studies identified 30 + genetic variants associated with TL (Codd et al., 2013; Li et al., 2020a). These findings have led to increased knowledge, and many studies have provided evidence for causal associations between short leukocyte TLs and age-related diseases (Kuo et al., 2019). Hence, it seems that the biology of telomeres is a good example of how genes and the environment interplay to present a phenotype. Genetic liability contributes to the overall length of telomeres in all cells, and across the lifespan, stressors and lifestyle factors influence cell-specific attrition rates. Different aging theories may fit in this scenario, while the limit on cellular division (Hayflick) was described as a direct consequence of critically short telomeres.

The sexual dimorphism of telomere dynamics has been discussed in many different aspects (Barrett and Richardson, 2011). The sex chromosome-linked mechanisms could be part of the explanation. Although most telomere-related genes have been found in autosomal chromosomes, it has been suggested that the unguarded chromosome in heterogametic sex is a disadvantage for mortality and telomere maintenance. A mutation in the DKC1 gene on the X chromosome – a gene involved in telomere biology – is often seen in patients with dyskeratosis congenita, which leads to rapid TL shortening and reduced survival (Savage and Bertuch, 2010). Another explanation is that the larger sex has disadvantages in the cellular maintenance, oxidative stress reactions, and telomere function because cellular capacity is linked to growth. Consequently, men, who are generally taller than women, should suffer from worse telomere function. However, a recent meta-analysis investigated sex differences in TL across 51 vertebrate species and found no evidence supporting either the heterogametic sex disadvantage or the sexual selection hypotheses (Remot et al., 2020). The analyses, including TL dynamics in mammals, birds, reptiles, and fish, did not find associations to support sex differences in longevity. Hence, the true nature by which TL sexual dimorphism presents remains to be elucidated. The importance of sex hormones may need further scrutiny, as they influence the level of ROS production, which may interfere with telomere maintenance and elongation (Coluzzi et al., 2019). However, other theories have been discussed, and many factors are likely important for sex-specific telomere dynamics.

Cellular senescence
Another hallmark of aging is cellular senescence. The lifetime of a cell is limited, as described by Hayflick, and the fate of a cell depends on the type of cell and what signals it receives and the damage it is exposed to across life. Events such as critically short telomeres, oxidative stress, replicative errors, mitochondrial dysfunction, pathogen response, oncogene activation, and other stress sources may induce senescence of the cell with irreversible replicative arrest (López-Otín et al., 2013). This state causes a response of cytokines and other proinflammatory factors to be released, which may trigger downstream effects in the surrounding tissue and invoke a senescence-associated secretory phenotype (SASP) (Ferrucci et al., 2020). Cellular senescence is tightly linked with aging, has been well correlated with DNA damage, and an increasing number of cells are senescent in old tissues compared to young tissues in a study of liver tissue in mice (López-Otín et al., 2013; Khosla et al., 2020). However, it has been difficult to assess SASP in human studies since the phenotype markers are heterogeneous and not consistently available in the circulation. Nevertheless, the systemic accumulation of senescent cells in aging has been associated with many age-related diseases and conditions, such as frailty, both in humans and animal models (Ferrucci et al., 2020; Khosla et al., 2020; Schafer et al., 2020). Currently, there is also increasing evidence for the beneficial antiaging effect of senolytic drugs as potential treatments to remove senescent cells when abundant (Ferrucci et al., 2020). Hence, cellular senescence is a core mechanism in the senescence theory of aging, where cells and tissues accumulate damage across life but is also essential in the Hayflick limit's programmed theory of aging (Ferrucci et al., 2020; Khosla et al., 2020; Schafer et al., 2020).

No human studies specifically investigate the difference between men and women in cellular senescence, and evidence from other models is sparse. A recent study in mice suggested that male mice have a higher number of senescent cells across life compared to female mice (Yousefzadeh et al., 2020), although at the end of life, the proportion of female senescent cells is almost at the same level as in male mice. The notion of higher cellular senescence in males would be consistent with the shorter telomeres seen. Evidence points to the fact that female stem cells have an increased capacity for regeneration, self-renewal, and proliferation (Dulken and Brunet, 2015), in line with a more beneficial cellular aging route in females/women. The limited knowledge would nevertheless suggest that sexual dimorphism exists, where women maintain better cellular maintenance throughout the life course. Regardless, more studies on sex differential senescent mechanisms are urgently needed to learn about the biological aging processes therein.

Proteostasis and autophagy
Protein homeostasis, or proteostasis, is the body's ability to maintain control over protein synthesis, folding, stability, degradation, and removal through autophagy (Hipp et al., 2019). During aging, the balance in the protein machinery is lost and unfolded and misfolded proteins can aggregate and cause pathological conditions seen in diseases of (neuro)degeneration, AD, PD, and diabetes (Hipp et al., 2019). Oxidative stress and heat may increase conformational changes and induce cellular toxicity from accumulated protein aggregations. Under stressful conditions, the heat shock response is activated in the cell, and unbound chaperones are available to assist in stabilizing the protein network. A study by Ubaida-Mohien et al. found a decreased representation of chaperone proteins in old skeletal muscle tissue in healthy adults, although autophagy-related proteins were overrepresented (Ubaida-Mohien et al., 2019). Experiments in worms, flies, and mice have shown that overexpressing chaperones and heat-shock proteins are associated with an extended lifespan, whereas models deficient in parts of the chaperone-heat-shock system present accelerated aging phenotypes (López-Otín et al., 2013; Ferrucci et al., 2020). Moreover, autophagy becomes dysfunctional with aging. In model systems, abrogation of autophagy leads to neurodegeneration and shortens lifespan, whereas increased basal activity of autophagy increases lifespan (Leidal et al., 2018). In humans, long-lived families have a better-maintained autophagy system, and individuals under starvation exhibit enhanced autophagic flux (Leidal et al., 2018). Hence, declining proteostasis control in aging may be an effect of accumulated aggregates and dysfunctional autophagy, consistent with the senescent wear-and-tear theory of aging, including the ROS theory of aging.

A recent investigation analyzed proteasome activity across nine different tissues and found higher activity in female mice than in their male counterparts (Jenkins et al., 2020). The largest sexual dimorphism was observed in the small intestine and kidney, specifically in chymotrypsin-like proteasomal activity. In another study, female fruit flies were more tolerant to oxidative stress and showed increased proteasome expression and activity than male flies, although the resistance was lost with age (Pomatto et al., 2017). Overall, adaptations to maintain homeostasis seem to depend on both age and sex, although studies on the latter are still limited (Pomatto et al., 2018). Females studied in animals and model systems also exhibit more resistance to stressors, partly hypothesized to be due to estrogens' beneficial effects (Tower et al., 2020). Analyses on sexual dimorphism in human protein homeostasis and autophagy aging processes are still lacking, which is understandable, as efficient high-throughput methods are not yet available (Ferrucci et al., 2020; Pomatto et al., 2017).

Epigenetic alterations
The term ‘epigenetics’ means ‘on top of genetics’ and is a collective term for chemical modifications altering the activity of the gene transcription process without changing the DNA code itself. There are four major types of epigenetic mechanisms: ATP-dependent chromatin remodeling complexes, histone, and DNA modifications, and noncoding RNAs (Pagiatakis et al., 2021). Histones can be modified posttranslationally. The most well-studied mechanisms are acetylation and methylation processes; changes in histone acetylation/methylation have been linked to aging, healthspan, and lifespan in diverse models, such as flies, mice, yeast, and human cell lines (Yi and Kim, 2020). A study by Klein et al., 2019 found that tau may affect histone acetylation in the human brain using an epigenome-wide association study of H3K9ac, thus relating histone modification processes to AD pathology. However, in human studies, genome-wide DNA methylation arrays have paved the way for a new field of research on epigenetic age where hundreds of (un)methylated sites (CpGs) have been shown to associate with age across the life course (Zhang et al., 2020). A multitude of clocks quantifying biological age across tissues, in whole blood, skin, muscle, or in human cell culture models have emerged (Horvath and Raj, 2018) and recently across mammalian species (Lu, 2021). With remarkable accuracy, clock ticks with aging and a higher epigenetic age are associated with worse health and increased mortality risk (Horvath and Raj, 2018; Chen et al., 2016). Promising studies have reported reversal of epigenetic age with different interventions (Horvath, 2020; Fahy et al., 2019). A still unanswered question is whether this reversal of the epigenetic clock would then infer a lower risk for adverse events. In other words, is the epigenetic process causal in aging (Zhang et al., 2020)? As with telomeres, the epigenetic clocks seem to be tightly linked with cellular replication underlying the Hayflick limit theory (Wagner, 2019). Genetic studies of epigenetic clocks have discovered several loci associated with lifespan and lifestyle factors beyond the gene regions where the CpGs themselves are located (Lu et al., 2018; McCartney, 2020). One of the top loci found harbors the telomerase TERT gene, demonstrating the link to telomere biology. Epigenetic changes have also been proposed due to both developmental and maintenance processes, where gestational age clocks represent the former and other adult tissue clocks represent the latter. Moreover, intrinsic and extrinsic epigenetic clocks have been suggested to represent internal (cellular) versus external (lifestyle stressor) aging processes (Horvath and Raj, 2018). The epigenetic process in aging may be consistent with both senescence and programming theories on aging depending on the specific timing in life and the clock under study.

The sex-specific effect on epigenetic age is apparent in young children and adults (Horvath et al., 2016; Horvath and Raj, 2018). At all ages, boys/men have a higher epigenetically predicted biological age than girls/women, in accordance with the survival benefit in women. This phenomenon seems to be true across different tissues and gives rise to an effective difference in mortality risk between men and women (Li et al., 2020b). Moreover, in women, earlier menopause, either natural or surgical, is associated with increased epigenetic age, and although the finding was not consistent across different tissues, there was further support for lower epigenetic age in women undergoing HRT (Levine et al., 2016). Little is known about sex-dimorphic effects on histone modifications in aging, although studies on different interventions and acetylation/methylation in animals suggest that these effects are important modifiers in aging (Fischer and Riddle, 2018). Furthermore, the Klein study found >4000 H3K9ac sites associated with sex in their human histone data, highlighting the future need for deeper studies in this area (Klein et al., 2019). Studies investigating genome-wide DNA methylation differences between men and women report significant differences in autosomes and on the X chromosome, the latter being linked to sexual dimorphism genes and XCI (Li et al., 2020c; McCartney et al., 2019). A recent meta-analysis study investigating the age-related sex differences in DNA methylation patterns found changes associated with both methylation level and variability across the genome (Yusipov, 2020). Differentially methylated sites were enriched in imprinted genes but not in sex hormone-related genes. Furthermore, the top CpGs displayed a sex-specific pattern in samples from centenarians (healthy aging model) and Down's syndrome (accelerated aging model). On the other hand, another study investigating brain DNA methylation patterns found no support for sex-age interaction effects in neurodegeneration from human samples on AD and controls (Pellegrini, 2020). Studies on sexual dimorphism and DNA methylation are sparse in animal models, but some evidence for differences has been found in both rats and mice (Sampathkumar et al., 2020). Bacon et al., 2019 used a rat model resembling human neuroendocrine function and showed that DNA methylation regulates the onset of menopause. Taken together, the sexual dimorphism seen in epigenetic studies on aging is complex and seems to reflect sex chromosome-linked mechanisms and/or hormonal biological processes.

Inflammatory and immunological makers
Immunoinflammatory functions are at the heart of health in aging, and there is exhaustive literature available on the various changes that take place with age. At a cellular level, two distinct yet often parallel processes characterize immune aging: immunosenescence and inflammaging. The former refers to changes in the adaptive immune system, such as increased numbers of memory CD8 +T cells (resulting in a decreased CD4/CD8 cell ratio), loss of the key costimulatory molecule CD28 on the T cell surface and compromised clonal expansion and specific antibody production in the B cell compartment (Gubbels Bupp, 2015; Franceschi, 2019). Inflammaging refers to chronic, low-grade inflammation that occurs in the absence of infection and manifests as increased production of proinflammatory cytokines, linked to both frailty and CVD (Ferrucci and Fabbri, 2018). From an evolutionary perspective, inflammaging can result from positive selection of genetic variants that associate with higher levels of pro-inflammatory factors and enhanced immune responses in early life, conferring better protection against pathogens but resulting in increased damage to host tissues in later life. Inflammaging is thus in accordance with multiple different theories, where various stimuli, such as oxidative stress and lifestyle factors, contribute as well (Franceschi, 2019; De la Fuente and Miquel, 2009).

While both sexes experience aging-associated changes in the immune system, the hallmark features differ for men and women, and men are considered to experience maladaptive changes to a greater extent (Gubbels Bupp, 2015; Gomez et al., 2018). Between puberty and menopause – when differences in the hormonal milieu are the greatest between men and women – women experience lower rates of infections, an advantage attributed to stronger immune and vaccine responses and more efficient pathogen clearance (Gubbels Bupp, 2015). On the other hand, women are more susceptible to autoimmune diseases than men. However, after the age of menopause, the incidence of autoimmune diseases in women decreases close to the numbers observed in men, whereas the incidence of chronic inflammatory diseases increases (Gubbels Bupp, 2015). The temporal dynamics of these changes point to the crucial role of sex hormones in shaping immune aging, although it is likely much more complicated, involving an interplay of multiple homeostatic systems. It has been shown that nonimmune cells, such as adipocytes, fibroblasts, and endothelial cells, also contribute to inflammaging (Franceschi, 2019). As stated above, men seem to experience immunosenescence to a greater extent than women, potentially because women exhibit higher basal immunoglobulin levels, higher CD4 +T cell counts, and an increased CD4/CD8 T cell ratio compared to men (Gubbels Bupp, 2015; Gomez et al., 2018). The corresponding adaptive immune functions, such as antigen-specific antibody responses and CD4 +T cell cytokine production, are also typically more enhanced in women (Gubbels Bupp, 2015; Gomez et al., 2018). A recent study using sequencing and flow cytometry data in blood mononuclear cells further elucidated the sexual dimorphism in immune aging by showing that male and female cells also significantly differ at the age when sex hormones decline (Márquez et al., 2020). Older women had higher genomic activity for adaptive immune cells, while older men had higher activity for monocytes and inflammation, indicating greater inflammaging in men (Márquez et al., 2020). In the same study, a life-course analysis of the timing of epigenomic regulation of chromatin accessibility showed that male immune cells are more strongly affected and that a decline in immune function occurs 5–6 years earlier in men than in women (Márquez et al., 2020).

Although animal models cannot fully recapitulate human immunosenescence or inflammaging, findings on sex-related immune functions in animal studies have generally been in line with observations in humans. Sex differences are present in diverse species ranging from insects to mammals, with female individuals presenting stronger innate and adaptive immune responses than males (Klein and Flanagan, 2016). Like humans, the differences are largely attributable to the effects of sex hormones, with a contribution of genetic differences due to several immunoinflammatory genes that are X chromosome encoded (Klein and Flanagan, 2016). In summary, the above findings support the assertion that men experience faster and/or earlier aging-associated immunoinflammatory changes and that these changes may be attributed to both hormonal changes and other factors.

Nutrient sensing
Intracellular nutrient-sensing pathways and signaling systems mediate information on nutrient availability and energy levels in the extracellular milieu. The key pathways include the insulin/insulin-like growth factor 1 (IGF-1) signaling pathway, mechanistic target of rapamycin (mTOR), and adenosine monophosphate-activated protein kinase (AMPK) pathway (Pignatti et al., 2020). These pathways regulate a multitude of intracellular functions, such as cell cycle control, DNA replication and repair, autophagy, and antioxidant defenses, by which their effects are excreted for reproduction, growth, and aging (Pignatti et al., 2020). Deregulated nutrient sensing is also one of the hallmarks of aging (López-Otín et al., 2013). Each of the hallmarks of aging is associated with undesirable metabolic alterations (López-Otín et al., 2016), stressing the fact that nutrient sensing and metabolism are interlinked processes with broad effects on whole-organism functions. Over the past years, there has been intensive research on how nutrient-sensing pathways control lifespan and healthspan, with the most significant breakthroughs achieved in unraveling how different dietary restrictions improve aging outcomes and survival in several species, including humans (Templeman and Murphy, 2018). Of the different dietary restrictions, the most compelling evidence rests on caloric restriction (CR), in which the energy intake is reduced ~30% relative to ad libitum-fed animals without reducing the intake of micronutrients (Templeman and Murphy, 2018). At the molecular level, CR triggers activation of stress response pathways that in turn reduce inflammation and increase repair and antioxidative functions. Interestingly, genetic polymorphisms in genes encoding proteins in the insulin/IGF and mTOR pathways are among those that are robustly associated with longevity, such that variants associated with the lower basal activity of the pathways are associated with longevity (Pan and Finkel, 2017).

Sex hormones regulate several key functions in nutrient sensing and metabolism of glucose, amino acids, and proteins, and it is not surprising that men and women differ in several metabolic characteristics. At the molecular level, women have lower fasting insulin and glucose levels, lower basal fat oxidation, and higher fat use but lower consumption of carbohydrates during physical activity (Comitato et al., 2015). The most noticeable difference is the fat distribution at the phenotypic level, so that men tend to have more visceral fat, whereas women have greater fat deposition in lower body depots (Comitato et al., 2015). For healthspan, the above-described traits tend to favor women such that they have a lower risk of cardiometabolic diseases (before menopause). However, the higher basal insulin levels in men promote glycogen and lipid synthesis in muscle cells, resulting in higher muscle mass and strength (Comitato et al., 2015). Aging is, however, associated with a reduction in glucose tolerance in both sexes, increasing the risk of diabetes. There is a complex interplay between sex hormones and body composition for which data from in vivo studies and clinical trials remain inconclusive (Allan, 2014). Future studies will hopefully shed light on possible sex differences in CR in humans; thus far, the available data do not support (or allow) inferences on sexual dimorphism. However, studies in rodents have suggested that males may have a more robust response to CR than females (Kane et al., 2018), but the mechanistic bases are not understood.

Akin to epigenetic clocks (see Epigenetic alterations) that predict mortality independent of other risk factors, there have been attempts to create similar composite measures based on metabolites measured using different techniques (Jylhävä et al., 2017). For example, Hertel et al., 2016 created a 'metabolic age score' that was shown to be associated with mortality independent of chronological age and other risk factors. The score was robustly associated with chronological age in both sexes, the only significant sex difference being that the score was more strongly influenced by obesity in women than in men (Hertel et al., 2016). However, such studies on metabolomics scores have been much fewer than studies on epigenetic clocks, and the potential sex dimorphism in metabolic scores is less clear. In summary, the sexual dimorphism in nutrient sensing and metabolism is largely attributable to sex hormones and their downstream effects. The higher muscle mass coupled with a higher basal metabolic rate in men also aligns with the rate of living theory.

Functional measures
Functional measures relevant to aging and mortality are numerous. One of the most commonly used and strongest markers for human population-based estimation of death risk is a simple assessment of walking speed (Ganna and Ingelsson, 2015), yet other popular measures include grip strength, chair rise, lung function, vision, and an abundance of cognitive domains (Peiffer et al., 2010). Although it is well known that being physically fit translates to better health, maintaining higher muscle mass and strength requires spending more energy and a higher metabolic rate. Analogous to the Hayflick limit, the rate of aging theory posits that the total amount of energy expenditure per lifetime is finite and that excessive usage results in accelerated aging (Pearl, 2011). Although much debated (Lints, 1989), this theory is supported by the observations that long-lived mammals have low energy expenditure rates, while short-lived mammals have higher rates. Studies in aging humans have shown that those having higher basic metabolic rates are more likely to die than those with lower rates (Ruggiero et al., 2008).

It is well established that men do better in physical capability, measured as grip strength, walking, and stair climb, even after adjusting for total body weight and lean body mass (Peiffer et al., 2010). Upon menopause, the withdrawal of sex hormones negatively affects bone and muscle health in women, where women experience a greater reduction in bone mineral density than men. However, men have a steady decline in bone function across life, but the interaction between load and bone strength is better maintained in older men, and this phenomenon may explain the reason for fewer fractures seen in men (Seeman, 2001). Women have less skeletal muscle mass than men, but men have greater loss with aging, although different parts of the body may show different sex-dimorphic effects, and menopause accelerates the loss in women (Doherty, 2003). Sarcopenia affects both sexes but is clinically more important in older women who may live longer with the disability (Doherty, 2003). For age-related visual impairment, women report more eye problems than men (Li et al., 2011), and overall, healthy adult men seem to perform better on visual perception than women (Shaqiri et al., 2018). In contrast, hearing loss is more frequent in men and may start as early as in the thirties (Shuster et al., 2019). Sexual dimorphism is also apparent in animal models, and women seem to be protected from age-related hearing decline before menopause, as estrogen levels are directly linked to the hearing threshold. Lung function is strongly associated with age, and a decline in spirometry-based measurements of dynamic flow starts soon after lung maturation in young adults (Sharma and Goodwin, 2006). Sex-specific differences are seen across almost all respiratory structures and functions; women have smaller and anatomically different lungs than men, perform worse in breathing exercises, and sex hormones interact with lung and airway function during early developmental processes and aging (LoMauro and Aliverti, 2018). However, anatomical changes during aging to other organs may be advantageous to women. Cardiac remodeling due to aging is universal, but the decline in myocytes and systolic function are greater in males, both in humans and rodents (Keller and Howlett, 2016). Kidney function declines with aging, and men have a greater decrease in glomerular filtration rate, where women are most likely protected due to estrogens before menopause (Baylis, 2009).

A recent study created a composite measure, termed the functional aging index (FAI), to better capture the state and changes in various physical functions simultaneously. The FAI includes muscle strength (grip strength), movement (gait speed), sensory (vision and hearing), and lung function and is predictive of mortality in both sexes, yet the hazard ratio is greater in women (Finkel et al., 2019). However, while women had higher FAI scores than men, indicating poorer functioning, the rate of change did not differ between the sexes (Finkel et al., 2019). Hence, the better physical performance in men may be explained by evolutionary selection for physical fitness, which means better health in general, but it is unclear why this does not translate to a survival advantage. As men have higher muscle mass than women, some clues might be obtained from the observed associations between higher skeletal muscle mass and higher basal metabolic rate, that is energy expenditure that is higher in men than in women (Ruggiero et al., 2008). Perhaps, the sex specificity in functional measures best describes the complex interplay between fitness and aging in line with the rate of living theory in the senescence theory of aging, emphasizing the sex paradox in aging where women with worse physical function and health still outlive men, possibly due to a better cellular maintenance system and protections from estrogens.

Frailty
Frailty is defined as a state of increased vulnerability to stressors resulting from decreased physiological reserves to maintain homeostasis across multiple organ systems. Manifestations of frailty overlap with those of normative aging yet are more pronounced. When a certain threshold in frailty is reached, the risk of adverse outcomes, such as disability and death, increases. Although frailty often coexists with multimorbidity (and disability), the association between frailty and mortality is independent of multimorbidity (Hanlon et al., 2018), indicating that frailty captures health-related variation that is not attributed to diseases alone. There is currently no widely accepted consensus on how to measure frailty; however, the two most commonly used approaches are the Fried phenotypic model (FP) (Fried et al., 2001) and the Rockwood frailty index (FI) (Searle et al., 2008). The first views frailty as a physical syndrome with a discrete categorization of individuals into nonfrail, prefrail, and frail, whereas the latter considers frailty as a multidimensional construct based on the accumulation of deficits in physical, biological, and psychosocial domains. The FI is measured on a continuous scale, allowing for the detection of more subtle changes and making the FI suited for younger individuals. Although viewed more as a measure of fitness than biological age, frailty stands out as an exception in the wealth of research devoted to understanding the sex differences compared to the other markers. Women not only have a higher prevalence of frailty but also experience higher levels than men across the age range (Gordon et al., 2017). Women are nevertheless able to tolerate frailty better; men are more vulnerable to death at any given level of frailty than women of the same age (Gordon et al., 2017; Jiang et al., 2017). The above-described male-female health-survival paradox may thus also be conceptualized as a sex-frailty paradox. The sex-frailty paradox has been described using several frailty scales (Theou et al., 2014) and across different populations (Gordon et al., 2017), suggesting that it is likely independent of the specific scale used to measure frailty.

The reasons for higher levels of frailty in women have been discussed previously, with various biological, social, and behavioral factors hypothesized to allow women to better tolerate frailty (Gordon and Hubbard, 2019; Hubbard, 2015). When conceptualizing frailty using the deficit accumulation model, that is the FI, it seems conceivable that women are evolutionarily ‘calibrated’ for late-life fitness. This theory aligns with the grandmother effect and increases in the population postreproductive lifespans when it benefits younger generations (Lahdenperä et al., 2004). Frailty also recapitulates characteristics of disposable soma theory that allow a certain amount of damage to the organism. However, another theory suggested underlying the sex differences is the chronic disease hypothesis by which women are more likely to experience nonlethal chronic conditions, while men tend to develop acute conditions associated with high mortality, such as stroke and myocardial infarction (Gladyshev, 2014; Bernabeu, 2020). Women may also be more prone to actively seek medical help for their conditions, resulting in better treatment balance of their (chronic) diseases. Last, variability in reporting behavior may contribute to the difference; when using self-reported data, a common conception is that men tend to underreport their morbidities and disability, while women are more likely to overreport. However, evidence supporting this conception is not conclusive (Merrill et al., 1997; Macintyre et al., 1999), and the underlying mechanisms for the sex-frailty paradox remain unresolved.

In recent years, animal models of frailty, building on both FI and FP, have become available, providing opportunities to untangle how and why frailty develops and the mechanisms behind the sex differences. However, evidence on sex differences in frailty in animal models is less equivocal than in human studies. Few studies have reported that aged female mice exhibit higher FI scores than males (Heinze-Milne et al., 2019). However, other studies have reported no difference between the sexes, and one study found that male mice had higher FI scores than females (Heinze-Milne et al., 2019). The paucity of animal studies available and the variety in mouse strains used in the studies nevertheless warrant more evidence before the mechanisms of the sex differences in frailty can be resolved.

Sex differences in age-related diseases
Due to global aging and improved health care, the leading causes of death worldwide have shifted remarkably over the last century. Noncommunicable diseases, which are considered chronic age-related illnesses, are now the three most common causes of death worldwide (ischemic heart disease, stroke, and chronic obstructive pulmonary disease) (World Health Organization, 2021). For the population older than 70 years, all but one (lower respiratory infection) of the top 10 leading causes of death in the world are noncommunicable age-related diseases (Table 1; World Health Organization, 2021). An age-related disease can be defined as a disease where chronological age is a strong risk factor, and the incidence rate is increasing with increasing age. For a more comprehensive review on age-related diseases and the link to biological aging mechanisms, we refer to Franceschi et al., 2018. However, age-related diseases often present in a sex-specific manner. The top 10 leading causes of death by sex in those above 70 years reveals a change in the ranking of diseases so that instead of colon and rectum cancers, prostate cancer emerges in men and communicable diarrheal diseases in women. Hence, we highlight the sexual dimorphism in age-related diseases below, further strengthening the evidence that biological aging is different in men and women.

Although men and women present different disease-specific patterns and expression of risk factors, several leading age-related diseases are related to cardiovascular health in both sexes. It is well accepted that premenopausal women are relatively protected from the most common cardiometabolic manifestations, whereas postmenopausal women are not (Aggarwal et al., 2018). This observation has been attributed to estrogens' beneficial effects on CVD, metabolic syndrome, and diabetes. (For a more in-depth discussion on sex and gender aspects in aging diseases and treatment, we refer the reader to Mauvais-Jarvis et al., 2020 and Regitz-Zagrosek, 2012). In addition to looking at the sex hormones individually, several studies have shown that it may instead be the sex-specific testosterone/estradiol ratio that is more decisive on health outcomes than either of the hormones alone (Morselli et al., 2016). However, CVD is also tightly linked to inflammaging of the vasculature, and cellular senescence that could be reflected as TL shortening and intrinsic epigenetic age acceleration (Ferrucci and Fabbri, 2018). Hence, aging and sexual dimorphism in cardiovascular health are delicately intertwined.

Most cancers have apparent sex-differentiated effects, even after controlling for risk factors and lifestyle differences between sexes. In general, men have higher incidence rates and higher death rates in most cancers that are not related to reproduction (Mauvais-Jarvis et al., 2020). The male predominance is seen already in children with cancer before puberty, indicating that genetic or early developmental processes going wrong likely determine these differences. All cancer tumors have mutations in their genome, and commonly mutated genes are referred to as oncogenes (Stewart and Wild, 2014). There are many oncogenes known across the genome, some with specific X-linked mutational differences in men and women, and others encoded by the Y chromosome. Recent evidence suggests that noncoding genomic regions also contribute to sexual dimorphisms in driving cancer mutations and signatures (Li et al., 2020d). Many oncogenes present specific epigenetic signatures used for cancer diagnostics (Stewart and Wild, 2014), and epigenetic outlier burden is associated with age and cancer diagnosis in a sex-specific manner (Wang et al., 2019). Longer telomeres and extrinsic epigenetic age acceleration are also features seen in cancerous tissues. Hence, genomic instability, including the accumulation of mutations, epigenetic alterations, and telomere attrition, are hallmarks of aging and provide a link between aging and sexual dimorphism mechanisms in cancer. There are also cancers related to hormonal secretion where androgens are stimulating and estrogens are protective (Hammes and Levin, 2019). Cancers are not a class of homogenous diseases but complex, age- and sex-dependent biological processes that may arise due to several different factors.

AD and other dementias are perhaps the most established age-related diseases, and the prevalence continues to grow worldwide because of global aging. They are predominant in women, particularly in the oldest old, which may also be attributed to the female survival benefit (Mauvais-Jarvis et al., 2020; Mazure and Swendsen, 2016). There is evidence for sex-specific brain differences in early growth and development of the brain and adult structure and function, which may be of relevance to neurodegeneration. Cognitive aging in healthy adults demonstrates sex-differential effects, where men generally perform better in visuospatial ability and women better in verbal ability, but the speed of decline may be worse in men, although the literature is not consistent (Li et al., 2020b; McCarrey et al., 2016). In AD, women present worse clinical symptoms for comparable levels of brain atrophy in men, and interactions with hormones may be one explanation for the differences (Toro et al., 2019). Early natural or surgical menopause and late initiation of HRT is associated with increased risk of AD (Mauvais-Jarvis et al., 2020). However, sex-differential effects may also be related to sex chromosomes. A recent study using an AD model in mice, expressing the human amyloid precursor protein, showed that adding an extra X chromosome decreased mortality and clinical AD symptoms (Davis, 2020). It should also be noted that sex differences in dementia incidence may be partially explained by selective survival (Shaw et al., 2021). Sex differences in the age-related diseases, frailty and domains of physical functioning are summarized in Figure 2.

Figure 2