The Multifaceted Roles of DNA Repair and Replication Proteins in Aging and Obesity

Abstract

Efficient mechanisms for genomic maintenance (i.e., DNA repair and DNA replication) are crucial for cell survival. Aging and obesity can lead to the dysregulation of genomic maintenance proteins/pathways and are significant risk factors for the development of cancer, metabolic disorders, and other genetic diseases. Mutations in genes that code for proteins involved in DNA repair and DNA replication can also exacerbate aging- and obesity-related disorders and lead to the development of progeroid diseases. In this review, we will discuss the roles of various DNA repair and replication proteins in aging and obesity as well as investigate the possible mechanisms by which aging and obesity can lead to the dysregulation of these proteins and pathways.

1. Introduction

Since the discovery of the double-helical structure of DNA over 50 years ago, an incredible effort has gone into understanding how genome stability is maintained. Thousands of DNA lesions occur per cell per day, resulting from both endogenous and exogenous sources. If left unrepaired or repaired in an error-generating fashion, then these lesions can lead to an accumulation of DNA damage and genomic instability. Multiple DNA repair pathways have evolved to remove various types of DNA damage from the genome. For example, exposure to ultraviolet light from the sun can lead to the formation of pyrimidine dimers, which are recognized and repaired by the nucleotide excision repair (NER) pathway[1]. Reactive oxygen species (ROS) generated from exogenous sources and metabolic processes within the cell can result in oxidative DNA damage that can be repaired by the base excision repair pathway (BER)[2]. DNA double-strand breaks (DSBs) caused by replication fork collapse or γ-irradiation, for example, are repaired by homologous recombination (HR) and non-homologous end-joining (NHEJ) pathways[3, 4]. Thus, the dysregulation or impairment of these DNA repair pathways can lead to an accumulation of DNA damage and/or mutations, resulting in genomic instability and the development of various genetic diseases.

Aging is a dynamic process by which various molecular processes decline over time, ultimately resulting in senescence and/or cell death. Common features associated with aging include graying hair, loss of skin plasticity, impaired organ function, and increased susceptibility to disease. Molecular hallmarks of aging include telomere shortening, early-onset of cellular senescence, increased DNA damage, decreased DNA repair, mitochondrial dysfunction, and loss of proteostasis (Figure 1)[5]. Mutations in several different DNA repair proteins can result in accelerated aging disorders (i.e., progeroid disorders), with the end result often being premature death. Many of these proteins and their involvement in preventing premature aging will be discussed in this review.

Figure 1.

Common hallmarks in aging and obesity. Characteristics frequently observed in aging (left) and obesity (right) are illustrated above. Hallmarks common to both aging and obesity are displayed in the overlapping section of the diagram and are referred to in the text. ROS, reactive oxygen species.

A number of theories have been proposed to explain the aging process[6]. The “telomere theory of aging” suggests that telomeres, a DNA sequence repeat that caps chromosome ends to protect the genome, shorten with each round of DNA replication and upon reaching a certain length the cells are unable to continue replicating, resulting in senescence and cell death[7]. Another theory, the “free radical theory of aging”, suggests that free radicals produced by endogenous processes attack various macromolecules, including proteins and DNA, resulting in an accumulation of DNA damage that with time overwhelms the repair capacity of the cell, resulting in cell death (Figure 2)[8]. It is likely that both of these events, among others, contribute to the overall process of aging, and thus understanding the underlying mechanisms involved is paramount in preventing premature aging and age-associated diseases.

Figure 2.

Schematic overview demonstrating how aging and obesity can lead to genetic instability. ROS, reactive oxygen species.

Another state that can influence genomic maintenance and stability is obesity. Obesity is most commonly characterized by an excessive increase in adipose tissue and chronic low-grade inflammation, which can lead to an accumulation of toxic by-products[9]. For example, individuals with obesity often have increased levels of ROS, which can oxidize DNA and lead to DNA damage[10]. Further, individuals with obesity have also been reported to have decreased levels of DNA repair, resulting in an accumulation of DNA damage, mutations, and genomic instability (Figure 2)[11].

In this review, we investigate how aging and obesity play roles in the dysregulation of DNA repair processes, as well as how defects in repair and replication proteins can exacerbate aging- and obesity-related diseases.

2. DNA Damage Repair

Genetic instability is a common underlying event in conditions of aging and obesity and plays a role in many associated disorders. Therefore, it is imperative that our cells have efficient mechanisms of DNA repair to ensure maintenance of genomic integrity. Several DNA repair processes have evolved to recognize and repair different types of DNA damage, including damaged or mismatched bases, single-stranded breaks (SSBs), and DSBs. In this section, we will discuss the roles of DNA repair proteins in aging- and obesity-related diseases and how defects in these pathways and proteins contribute to these diseases (see Table 1).

Table 1.

Roles and functions of DNA repair and replication proteins in aging- and obesity-related phenotypes.

Protein	DNA Repair Pathway	Role	Phenotypes				References
			Pre-mature Aging	Metabolic Dysfunction	Cellular Senescence	Telomere Dysfunction
OGGI	BER	Removes 8-oxoG lesions		✓		✓	[32,33,45,46,47,52,53,54,55]
NEIL Proteins	BER	Removes oxidized pyrimidines, oxidized cytosine derivatives, FapyA, 8-oxoA, and thymine glycol		✓		✓	[56,58,59,67,68,69]
APE1	BER	Recognizes AP site after removal of lesion and creates a nick for gap synthesis to occur			✓	✓	[72,73]
POLB	BER	Removes 5'dRP during short-patch BER and fills in the nucleotide gap left behind by APE1			✓		[74,75,76]
XRCC1	BER	Complexes with LigaseIII to seal in the nick via ligase activity and complete the BER process		✓			[79]
ERCC1-XPF	NER	Performs DNA incision at the 5'-end of the lesion	✓		✓		[84,85,86,90,91,97]
XPA	NER	Verifies lesion and recruits other repair factors	✓	✓			[97,100]
XPB	NER	Subunit of TFIIH and involved in unwinding of double helix after damage recognition	✓			✓	[97,103]
xpc	NER	Recognizes bulky lesion during global genome NER	✓			✓	[97,101]
XPD	NER	Subunit of TFIIH and involved in unwinding of double helix after damage recognition	✓			✓	[97,103]
XPG	NER	Performs DNA incision at the 3'-end of the lesion	✓		✓		[97,102]
XPV	NER	Functions as a translesion synthesis polymerase during lesion bypass	✓	✓	✓		[97,105,106]
CSA	NER	Recruits other NER factors during TC-NER	✓		✓		[107,108,110]
CSB	NER	Recruits other NER factors during TC-NER	✓		✓	✓	[107,108,111,113,114,115]
Ku70/80	NHEJ	Recognizes and stabilizes broken DNA ends	✓		✓	✓	[153,154]
DNA-PKcs	NHEJ	Phosphorylates NHEJ factors and induces a conformational change to allow access of broken ends to NHEJ proteins	✓			✓	[154]
Rad51	HR	Binds ssDNA and facilitates strand invasion into the sister chromatid				✓	[171,172,173,176]
BRCA2	HR	Mediates recruitment and assembly of Rad51 to ssDNA prior to strand invasion				✓	[172,173]
Rad54	HR	Interacts with and guides Rad51-ssDNA complex along sister chromatid dsDNA during homology search				✓	[175]
WRN	DNA Replication/DNA Repair	Functions as a DNA helicase to unwind the DNA for various metabolic process	✓	✓	✓	✓	[180,181,183,186,189,190,191,192,194]
BLM	DNA Replication/DNA Repair	Functions as a DNA helicase to unwind the DNA for various metabolic process			✓	✓	[201,202,203,204,205]
FEN1	DNA Replication/BER	Cleaves 5'-flap structures that arise during DNA replication and long-patch BER				✓	[217,218,222,223]
Twinkle	Mitochondrial DNA Replication	Functions as a mitochondrial DNA helicase to unwind the DNA at the replication fork		✓			[226,227,228,229]

2.1. Base Excision Repair (BER)

The BER pathway is a conserved and ubiquitous mechanism responsible predominately for the repair of single base, non-helix-distorting DNA damage (e.g., oxidative damage, deamination, alkylation, etc.). Initiated by various DNA glycosylases, the BER pathway is responsible for recognizing and repairing thousands of lesions per cell per day[2, 12]. Briefly, a lesion-specific glycosylase recognizes and removes the damaged base by cleaving the glycosidic bond between the damaged base and phosphate backbone, leaving an apurinic/apyrimidinic site (AP site). AP Endonuclease 1 (APE1) then makes an incision in the backbone generating 3’hydroxyl (3’OH) and 5’deoxyribosephosphate (5’dRP) ends[13]. In most cases, the 5’dRP moiety can be removed by polymerase β (pol β), which then uses the opposite DNA strand as a template to fill in the nucleotide gap. Finally, DNA ligase IIIα (LigIII) and X-ray cross-complementing protein 1 (XRCC1) form a complex to seal the remaining nick, thereby completing the short-patch BER pathway[14]. However, when pol β is unable to remove the 5’dRP (e.g., the moiety has undergone oxidation, reduction, etc.), then another sub-pathway of BER, long-patch BER, can process the damage. In long-patch BER, a single nucleotide is still inserted by pol β, but then polymerase delta/epsilon (pol δ/ε) add additional bases following the newly inserted base. The addition of several bases by pol δ/ε displaces downstream nucleotides, creating a 5’flap containing the 5’dRP, which is recognized and processed by flap endonuclease 1 (FEN1) and proliferating cell nuclear antigen (PCNA), leaving a nick in the backbone that is sealed by DNA ligase I (LigI) and PCNA[15-17]. An overview of this pathway is summarized in Figure 3. With over 20,000 base lesions occurring in each cell per day, BER is crucial for maintaining genomic integrity, and defects in this pathway can result in the development of several diseases, including cancer[18-21]. In fact, mouse models deficient in the BER pathway, particularly via pol β and APE1 mutations, are either embryonically lethal or die shortly after birth[22-24]. In this section, we will discuss several BER-related enzymes and their roles in aging and obesity.

Figure 3.

Schematic overview of DNA repair proteins and pathways discussed in this review.

2.1.1. 8-Oxoguanine DNA Glycosylase (OGG1)

OGG1 is a lesion-specific glycosylase in the BER pathway primarily responsible for the removal of 7,8-dihydro-8-oxoguanine (8-oxoG), a common lesion formed by ROS. If left unrepaired, this mutated base can mispair with adenine during DNA replication, generating a G:C to T:A transversion[25-28]. 8-oxoG is one of the most common oxidative lesions formed and, given its mispairing tendencies, is highly mutagenic[29, 30]. OGG1 is localized to both the nucleus and mitochondria. It has been reported that accumulation of mitochondrial DNA damage can lead to increased oxidative stress, mitochondrial dysfunction, and reduced insulin sensitivity[31]. Several studies have reported that mice deficient in OGG1 demonstrated a lower tolerance for cellular oxidative stress and were more prone to developing metabolic disorders[32, 33]. In this regard, OGG1-deficient mice displayed significant weight gain when fed a high-fat diet (HFD) and showed signs of fatty liver, including increased fat mass, lipid accumulation, triglyceride content, and insulin resistance, all signs of metabolic dysfunction[32]. These observations may be explained, at least in part, by reduced hepatic glycogen content, increased glycolytic gene expression, and increased respiratory exchange ratios in OGG1-deficient mice compared with wild-type (WT) mice when fed HFDs, all of which indicated decreased hepatic fatty acid oxidation[32]. These results suggested that reliance on fatty acids as fuel sources had decreased, while reliance on glycogen as a fuel source increased in the OGG1-deficient mice.

Interestingly, Scheffler et al. (2018) reported that OGG1-deficient mice on chow diets demonstrated increased expression of genes involved in gluconeogenesis in hepatic mitochondria compared to WT mice, contrary to the results from the mice on high-fat diets, where the expression of glycolytic genes was increased compared to WT mice[32, 33]. This discrepancy in gene expression could indicate different roles for OGG1 in regulating mitochondrial function during different energy states. In contrast to both of these studies, which demonstrated altered mitochondrial respiration and dysfunction, Stuart et al. (2005) found that mitochondrial respiration remained unaltered in OGG1-deficient mice[34]. It should be noted, however, that this group used mouse tissues pooled from various age groups, while the other studies used age-matched tissues, which may have impacted the results[32-34]. It has been well established that mitochondrial dysfunction is a characteristic of aging[35]. This could explain the lack of observable differences in mitochondrial respiration in OGG1-deficient and WT mice, further emphasizing the importance of sources of variation when utilizing mouse models.

These observed changes in OGG1-deficient mice are not limited to the liver, as another study reported that skeletal muscle tissue also presented with lipid accumulation, altered tissue metabolism, and decreased muscle function[36]. Several groups have also found polymorphisms in OGG1 in human patients that correlated with reduced insulin sensitivity, type 2 diabetes mellitus (T2DM), and obesity[37-39]. Additionally, the Lloyd and Sampath groups have reported in a collaborative study that overexpression of OGG1 actually protected mice against diet-induced obesity and suggested a role for OGG1 in maintaining mitochondrial energetics in white adipose tissue[40]. Together, these studies provide strong evidence for a role of OGG1 in modulating energy homeostasis on a whole-body level.

OGG1 has also been shown to be a regulator of age-related disorders. The “free-radical theory of aging” postulates that aging is largely the result of an accumulation of oxidative damage within the cell[41]. Consistent with this postulate, studies have shown that cellular senescence, a hallmark of aging, can be stimulated by high levels of oxidative stress[5, 42-44]. As one of the main enzymes involved in the repair of oxidative lesions, it can be hypothesized that OGG1 plays an important role in regulating the onset of aging and age-related diseases by facilitating the removal of the DNA damage caused by high levels of oxidative stress, thereby preventing the onset of cellular senescence. Indeed, several studies using various animal models have demonstrated increased levels of 8-oxoG and decreased levels of OGG1 with age[45-47]. Notably, many publications have reported that mitochondrial DNA, and not nuclear DNA, is where most of the oxidative damage is detected[28, 45, 47, 48]. Among many factors, this could be a result of: (1) nuclear DNA damage having a higher priority for DNA damage repair leading to lower levels of detected DNA damage compared to the mitochondria, and/or (2) the highly oxidative environment of the mitochondria, which could increase the susceptibility of mitochondrial DNA to oxidative damage compared to nuclear DNA. Regardless, further studies are required to address the mechanisms responsible for decreased 8-oxoG repair capacity in aging cells.

Another established theory of aging is the “telomeric theory of aging”[49]. Telomeres are comprised of DNA repeat sequences, (TTAGGG)_n, that act as caps to chromosome ends to prevent DNA damage responses (DDR) from recognizing them as broken DNA. Telomere shortening is a hallmark of aging and can lead to cellular senescence and apoptosis[5, 7]. Telomeric sequences are rich in guanines, making these regions susceptible to oxidative damage. Thus, it has been thought that one of the primary sources of telomere shortening in cells is oxidative stress[50, 51]. Rhee et al. (2011) reported that telomeric DNA repeats demonstrated higher levels of oxidized guanines than non-telomeric minisatellite TG repeats[52, 53]. Due to the repetitive nature of telomeric DNA sequences, studies have shown that telomeric regions are able to form alternative DNA structures, e.g., G-quadruplex DNA structures, which are formed in guanine-rich regions and susceptible to oxidative damage[54]. Because of this propensity for oxidative damage, DNA repair mechanisms are crucial in maintaining telomere integrity. In support of this, Wang et al. (2010) demonstrated that aged OGG1-deficient mice and OGG1-deficient mouse embryonic fibroblasts (MEFs) cultured in highly oxidative environments (i.e., 20% oxygen) accumulated higher levels of oxidized guanine in telomeric DNA than young mice or MEFs under environments of lower oxidation (i.e., 3% oxygen)[50]. In addition, OGG1 depletion in Saccharomyces cerevisiae resulted in significantly higher levels of oxidized guanine residues in telomeric DNA sequences, as well as altered telomere lengths[55]. These data indicated that OGG1 plays a role in maintaining telomere homeostasis and may act to prevent the premature onset of cellular senescence associated with aging. Regardless of which theory of aging is being considered, it is evident that OGG1 plays a crucial role in maintaining genome integrity and preventing the onset of several hallmarks of aging, including mitochondrial dysfunction, telomere attrition, and genomic instability.

2.1.2. Nei Endonuclease VIII-like (NEIL) Proteins

Belonging to the Nei endonuclease VIII-like (NEIL) family of glycosylases, the NEIL proteins, similar to OGG1, are involved in the removal of damaged bases to initiate BER. However, unlike OGG1, the NEIL proteins are able to recognize a broader range of damaged substrates. Currently, there are three reported NEIL protein glycosylases: NEIL1, NEIL2, and NEIL3. NEIL1 is primarily involved in the removal of oxidized pyrimidines, such as thymine glycol and 5’hydroxyuracil, in concert with the replication fork machinery before the DNA is synthesized by polymerases. NEIL2, on the other hand, preferentially removes oxidized cytosine derivatives, particularly in mismatched double-stranded DNA such as during transcription-coupled DNA repair. Less is known about NEIL3, but studies have shown that it can remove a wide variety of lesions in single-stranded DNA, including 4,6-diamino-5-formamidopyrimidine (FapyA), 7,8-dihydro-8-oxoadenine (8-oxoA), and thymine glycol, and that it can also remove oxidized bases from G-quadruplex structures[56, 57]. Due to their roles in BER, and the role of BER in regulating oxidative stress, it is thought that, like OGG1, defects in the NEIL glycosylases may lead to metabolic dysfunction and obesity. Indeed, NEIL1-deficient mice have been reported to be prone to obesity, hepatic lipid accumulation, hepatic inflammation, hyperleptinemia, hyperinsulinemia, and hyperlipidemia, all of which are indicative of metabolic dysfunction[58, 59]. In regard to its role in DNA damage repair, NEIL1-deficient mice also have increased mitochondrial DNA damage and deletions[58].

When placed on HFD, NEIL1-deficient mice displayed similar but heightened metabolic stress and dysfunction compared to mice on chow diets, including the addition of impaired glucose clearance, decreased mitochondrial content, and decreased lean body mass[59]. It is interesting to note that while many similar outcomes were observed in OGG1-deficient mice, alterations in lean body mass had not been reported, suggesting that glycosylase-specific activities in addition to disruptions in the BER pathway likely contribute to metabolic dysregulation. Additionally, polymorphic variants of NEIL1 resulting in defective glycosylase activity have been reported in human cancer patients; however, susceptibility of these patients to metabolic disorders due to these variants has not been reported[60, 61].

Unlike NEIL1, very few studies have reported on the potential roles of NEIL2 or NEIL3 in energy balance and metabolic regulation. From a limited number of studies, it was observed that NEIL3 was involved in the regulation of lipid metabolism in atherosclerosis-prone mice, and that several NEIL3 polymorphisms were associated with type 2 diabetes[62, 63]. Although the roles of NEIL glycosylases in obesity and metabolic syndrome have not been extensively studied, these reports provide a solid foundation for further research into the underlying mechanisms imparted by the NEIL proteins on metabolic homeostasis.

Requiring nearly 20 percent of the body’s oxygen supply, the brain occupies an oxygen-rich environment and can be prone to high levels of oxidative stress and increased susceptibility to oxidative DNA damage. Therefore, it is imperative that DNA repair mechanisms be rapid and efficient in their recognition and removal of oxidative lesions in the brain. It has been observed that with age, the levels of DNA damage increase in the brain. But whether this is due to increased accumulation of damage or decreased repair capacity, or a combination thereof, remains unclear. DNA glycosylases play major roles in the repair of oxidative lesions via their recognition and removal of oxidized bases. Reduced activity of these proteins could lead to inefficient DNA repair and a subsequent increase in the levels of DNA lesions remaining in the genome, resulting in genetic instability. NEIL1 and NEIL2 mRNAs are widely distributed throughout the brain, supporting a role for BER in maintaining genomic integrity in the brain[64]. Interestingly, studies have shown that NEIL activity in various regions of the brain does not decline with age, as one might expect; in fact, NEIL activity was shown to peak in middle-aged mice before tapering down in older mice with levels comparable to young and adult-aged mice[65]. Consistent with this, Rolseth et al. (2008) demonstrated that NEIL1 mRNA in the brain was increased in 12-month-old mice compared to 1 month-old mice[64]. In a NEIL1-deficient mouse model, Canugovi et al. (2012) reported that NEIL1 was crucial for memory retention, as mice deficient in NEIL1 consistently had lower performance levels in Morris water maze behavioral tests compared to WT mice[66]. These mice also exhibited increased apoptosis, brain damage, decreased DNA repair capacity, and decreased motor function after ischemic stroke, suggesting that NEIL1 may act to protect the brain against damage induced by severe stress. These results suggested that NEIL proteins, and NEIL1 in particular, played crucial roles in maintaining genomic integrity in the brain. Thus, additional investigations into the mechanisms of their regulation and function in the brain with age are warranted.

The NEIL proteins have also been reported to be involved in the repair of telomeric DNA damage. As a hallmark of aging, telomere integrity is critical in preventing the onset of cellular senescence and apoptosis. Zhou et al. (2013) determined that NEIL1 and NEIL3 removed lesions in G-quadruplex-forming sequences with a preference for lesions in telomeric DNA sequences. Furthermore, NEIL3 was reported to be recruited to telomeres during S phase in human cells via interactions with telomeric repeat factor 1 (TRF1) to remove oxidative DNA lesions prior to cell division, such that loss of NEIL3 resulted in telomere dysfunction[56, 67]. Although NEIL2 did not demonstrate any activity on telomeric four-repeat quadruplex DNA (TTAGGG)₄, NEIL2 did remove hydantoins from telomeric five-repeat quadruplex DNA (TTAGGG)₅[68]. Moreover, Chakraborty et al. (2015) reported that embryonic fibroblasts from NEIL2-deficient mice had greater telomere instability than the WT mice, suggesting that NEIL2, like NEIL1 and NEIL3, plays some role in telomere stability and maintenance[69]. This is further supported by evidence that NEIL2 processes lesions within D-loops, which have been reported to form at telomeres[70].

2.1.3. Other Proteins Involved in Base Excision Repair

Glycosylases, such as those mentioned above, initiate BER via the removal of damaged bases leaving AP sites that are processed by a network of proteins, including APE1, Polβ, XRCC1, etc. APE1, also commonly referred to as reduction-oxidation factor 1 (REF-1), recognizes the AP site and incises the DNA backbone so that gap synthesis can occur. APE1 is a unique protein in the BER pathway as it was independently identified as a redox factor and a DNA repair protein, hence the two commonly used names REF-1 and APE1, respectively[71]. Polβ, a well-studied BER protein, acts to remove the 5’dRP moiety left behind by APE1 and fills in the nucleotide gap generated by the removal of the damaged base. Mutations in Polβ are common in many different types of cancer, including colon, breast, gastric, and lung cancer. Indeed, results from small-scale studies suggest that Polβ mutants are present in at least 30% of human tumors, a staggering percentage attributed to just one protein[18]. After Polβ performs gap synthesis, the XRCC1-LigIII complex then seals the nick via DNA ligase activity, thus completing the repair process.

Of the proteins involved in BER, only the DNA glycosylases have been thoroughly studied for their roles in obesity and metabolic syndrome. While the BER pathway itself is critical for repairing oxidative damage, particularly in high states of oxidative stress such as obesity, it has yet to be determined whether deficiency of proteins downstream of DNA glycosylases contribute to obesity or metabolic phenotypes. However, complete knockout of APE1 results in embryonic lethality, suggesting that one or more of its roles is essential for embryonic development[22]. Moreover, APE1 knockdown in human primary cells and tamoxifen-induced knockout of APE1 in mice resulted in induced telomeric DDR, telomere shortening, and cellular senescence, suggesting a role for APE1 in preventing aging-related phenotypes[72, 73].

Similar to APE1, studies have demonstrated that mice harboring mutations resulting in the inactivation of Polβ are not viable. For example, Gu et al. (1994) found that Polβ-null mice died during embryogenesis, sometime after embryonic day 10.5 (E10.5), indicating that Polβ knockout was embryonically lethal[23]. Sugo et al. (2000), however, found that Polβ-null mice did not die during embryogenesis, but rather shortly after birth, likely due to a respiratory defect[24]. This discrepancy could be due to the nature of the Polβ gene mutation resulting in the Polβ-null phenotype. Another possibility is that because the mice died so soon after birth, the mother disposed of the deceased pups before they were noted. In any case, it is apparent that Polβ is critical for early-life development. Interestingly, mice heterozygous for Polβ showed increased chromosomal abnormalities, increased single-stranded DNA breaks, and appeared to die at a faster rate than WT mice[74, 75]. Additionally, loss of Polβ induced cellular senescence in mouse primary cells[76]. These studies suggested that, like APE1, Polβ plays a crucial role in the onset of aging.

Although no direct studies have been conducted to investigate the mechanisms linking XRCC1 and obesity, correlation studies in humans have determined that polymorphisms in XRCC1 are associated with obesity and Type 2 diabetes[77, 78]. Further, XRCC1 deficiency in testis of mice induced oxidative stress, mitochondrial injury, and mitochondrial dysfunction[79]. Together, these results warrant further studies into the potential role(s) of XRCC1 in obesity and metabolic syndrome. Like APE1 and Polβ, deletion of both alleles of XRCC1 is embryonically lethal[80]. The observation that complete knockouts of a number of proteins in the BER pathway resulted in non-viable mice demonstrated the importance of BER in maintaining genomic integrity during early life development. Unlike APE1 and Polβ, however, no obvious links between XRCC1 and aging have been reported. In fact, mice heterozygous for XRCC1 have the same median life span as that of WT mice[81]. This is an interesting finding, since it is known that deficiencies in APE1 and Polβ resulted in several hallmarks of aging (i.e., genetic instability, senescence, telomere length alterations, etc.), and that these proteins are all crucial for functional BER. That XRCC1 deficiency does not result in aging phenotypes may suggest alternative mechanisms for completing the repair pathway in the absence of this scaffold protein.

2.2. Nucleotide Excision Repair (NER)

NER recognizes and removes bulky, helix-distorting lesions within the DNA, typically caused by UV radiation or chemical agents (e.g., Benzo(a)pyrene, cisplatin, etc.). There are two sub-pathways of NER, transcription-coupled NER (TC-NER) and global genome NER (GG-NER). The primary difference between the two sub-pathways lies in how they recognize the DNA damage. In TC-NER, stalling of RNA polymerase II during transcription leads to the recruitment of TC-NER specific enzymes, including Cockayne Syndrome groups A and B (CSA and CSB, respectively), which act to recruit other factors required for NER. In GG-NER, the damage is recognized by the distortions induced in the double helix by the damage via UV-damaged DNA-binding protein (UV-DDB) and Xeroderma pigmentosum complementation group C (XPC). From this point onward, the repair process is similar for both sub-pathways. Briefly, XPA is recruited to verify the damage as RPA binds the undamaged strand. Transcription Factor II H (TFIIH) is then recruited to assist in the formation of an open complex that is a substrate for the endonucleases, ERCC1-XPF and XPG, which make incisions at the 5’ and 3’ ends of the damage, respectively, resulting in an excised DNA fragment containing the lesion. Replication factor C (RFC), PCNA, Polε/δ, and DNA ligase are then recruited to fill in the gap left behind, completing the repair process (Figure 3)[82], There are a limited number of studies that have investigated obesity and NER; however, numerous studies have determined that defects in NER result in several progeroid syndromes (i.e., premature aging).

2.2.1. Excision Repair Cross-Complementation Group1-Xeroderma Pigmentosum Group F (ERCC1-XPF)

ERCC1-XPF is a heterodimeric endonuclease that is involved in a number of DNA repair processes, including its essential role in NER. Within the NER pathway, ERCC1-XPF acts by making an incision at the 5’ end of the bulky lesion, contributing in the excision and removal of the lesion from the DNA[83]. Though rare, mutations in the genes encoding XPF and ERCC1 have been reported to lead to cellular senescence and accelerated aging[84-86]. To date, only two patients have been reported to contain non-synonymous mutations in ERCC1, demonstrating the lethality of variants in this gene[87, 88]. Both patients had impaired NER capabilities, severe developmental symptoms, and decreased lifespan. ERCC1 knockout mice are viable for a period of time, but experience significance growth retardation, liver failure, and die around weaning age[89]. ERCC1 deficient mice, with one allele deleted and the other hypomorphic, are viable for a longer period of time and demonstrate phenotypes of accelerated aging, making them a valuable model for aging studies[90]. ERCC1-XPF deficiency resulted in accumulated DNA damage, enhanced senescence-associated secretory phenotype (SASP) factors, cellular senescence, and apoptosis in human primary cells and mouse tissues[85, 90, 91]. Almost counterintuitively, however, ERCC1-XPF has been reported to be a negative mediator in telomere length maintenance, facilitating telomere shortening via TRF2-associated mechanisms[92, 93]. This is interesting, as telomere shortening is a hallmark in aging and, as described previously, deficiency in ERCC1-XPF leads to senescence and accelerated aging. This may suggest that the mechanism(s) by which ERCC1-XPF deficiency leads to accelerated aging is independent of telomere shortening. Further studies investigating ERCC1-XPF and telomere maintenance are warranted to shed more light on this intriguing phenomenon.

Along with an accumulation of senescent cells, ERCC1-XPF-deficient mice also experience increased oxidative DNA damage and susceptibility to lipid peroxidation (LPO)-induced DNA damage[94, 95]. Dietary fats, particularly polyunsaturated fatty acids (PUFA), are a source for endogenous LPO, and LPO by-products have been reported to be increased in obese individuals[9]. When placed on diets high in PUFA, ERCC1-XPF-deficient mice showed decreased maximum lifespan and age-related degenerative markers in their livers and kidneys[95]. Interestingly, while mice on the PUFA diet had increased glycogen storage, there was no weight gain when compared to mice on the control diet, suggesting that alterations in metabolic homeostasis may be independent of caloric intake. Additionally, Karakasilioti et al. (2013) demonstrated that ERCC1-XPF deficiency induced lipodystrophy, which can lead to further metabolic complications and abnormalities, further supporting a role for ERCC1-XPF in metabolic regulation[96].

2.2.2. Xeroderma Pigmentosum (XP) Complementation Proteins

A total of eight XP complementation groups have been identified thus far, ranging from XPA to XPG and XPV. Mutations in these genes can result in the autosomal recessive disease XP, a skin condition that causes the patient to be highly sensitive to UV irradiation, increasing susceptibility to skin cancer, and premature aging. Conditions of XP range from mild to severe symptoms, depending on the mutation[97]. Mutations in the first seven XP groups, XPA-XPG, often lead to defects in the NER mechanism. The eighth gene, XPV, encodes for the translesion synthesis protein, polymerase eta (polη)[98]. The XP proteins have diverse roles in DNA repair ranging from damage recognition, validation, DNA excision, and lesion bypass during DNA replication[99]. While their roles are diverse in nature, they are all central in preventing DNA damage-induced genetic instability.

As mutations in these genes lead to UV sensitivity and premature aging, it stands that efficient repair of UV-induced DNA damage is critical in preventing the onset of premature aging. Indeed, Fang et al. (2014) found that across several species, including mice, rats, and humans, XPA deficiency resulted in mitochondrial dysfunction and degradation, a prominent hallmark of aging[100]. Additionally, XPC-deficient MEFs displayed increased telomere fragility compared to WT MEFs, and upon chronic UV exposure had increased telomere shortening compared to WT[101]. Harada et al. (1999) observed that mice deficient in XPG failed to mature properly after birth and underwent premature death around weaning age. They also observed that primary cells from these mice displayed an accumulation of p53 and premature senescence[102]. Deficiency in the NER DNA helicases, XPB or XPD, subunits of THIIH, increased the sensitivity of cells to oxidative DNA damage, particularly at telomeres, and resulted in telomere attrition[103]. This is an interesting outcome, as it supports the hypothesis that UV-induced DNA damage alone does not account for the various phenotypes observed in XP patients, including effects seen on organs that are not exposed to UV irradiation. Together, these studies support the idea that efficient repair of DNA damage is crucial to prevent the premature onset of aging. However, further studies are needed to uncover the underlying mechanisms of how these proteins are involved in preventing premature aging, particularly their roles in telomere maintenance, as deficiencies in several of these proteins result in telomere attrition.

Because XP patients have a predisposition for cancer development and premature aging, much of the published work investigating mechanisms of XP proteins focus on their roles in carcinogenesis and aging; very few studies have evaluated the potential role(s) for the XP proteins in metabolic homeostasis. XPB, a subunit of TFIIH that is canonically involved in the unwinding of the DNA double helix after damage recognition, was found to interact with the BER protein OGG1 in the hippocampus of cholesterol-fed rabbits[104]. As this interaction did not occur in rabbits on a normal diet, it supported the idea of cross-talk between DNA repair pathways in response to cholesterol-induced lipid peroxidation and oxidative stress, a phenomenon also observed during obesity[104].

Interestingly, in another studying investigating polη-deficiency, mice deficient in polη displayed increased weight gain with age compared to WT littermate mice independent of caloric intake or water consumption[105]. Upon further investigation, the mice also displayed increased body fat percentage, hyperinsulinemia, reduced glucose tolerance, hepatic steatosis, and adipocyte macrophage infiltration when compared to WT mice, all of which are characteristics of obesity and metabolic syndrome. Additionally, polη-deficient mice showed increased levels of DNA damage and cellular senescence; interestingly, the cellular senescence was detected as early as 4 weeks of age, prior to the onset of obesity in the mice[105]. Cellular senescence can lead to adipocyte hypertrophy and lipid accumulation, two primary characteristics of obesity[106]. This suggests that DNA damage-induced senescence may have played a causative role in the onset of obesity in polη-deficient mice. Altogether, this study suggests a role for polη in regulating metabolic homeostasis, likely via its function in lesion bypass to maintain genomic stability and prevent DNA damage-induced senescence, providing further evidence of a role for genome integrity in obesity and metabolic syndrome. However, additional studies are needed to corroborate these findings and further explore the underlying mechanisms linking DNA damage, repair, and obesity.

2.2.3. Cockayne Syndrome (CS) Proteins

Similar to XP, CS is another progeroid disorder that occurs in individuals lacking either CSA or CSB, and is characterized by short stature, microcephaly, neurodegeneration, UV sensitivity, and accelerated aging[107, 108]. Deficiency of CSA or CSB can lead to defects in TC-NER, which is crucial for repair of DNA damage in actively transcribed regions. After DNA damage recognition by transcription polymerase stalling, CSA and CSB are required for recruitment of NER factors, such as TFIIH, as well as proteins for transcription restart after the damage has been repaired[109]. Due to this, defects in CSA and CSB, and therefore defects in TC-NER, can lead to transcription blockage and persistent DNA damage, resulting in genomic instability and premature cell death. Human primary keratinocytes deficient in CSA displayed abnormal secretions of SASPs, an abundance of oxidative damage, and cellular senescence[110]. Re-expression of WT CSA rescued these functions, indicating a protective role for CSA in maintaining genomic integrity and cellular function to prevent the onset of premature aging.

CSB expression has been reported to decline in human cells and mouse brain tissue with age; additionally, CSB-deficient cells experience heterochromatin loss, a phenomenon typically seen in aging cells[111, 112]. Chrochemore et al. (2019) reported that downregulation of CSB was an early event in replication-dependent senescence and that depletion of CSB induced p21-dependent cellular senescence[111]. In addition to preventing heterochromatin loss and cellular senescence, CSB appears to play a role in telomere maintenance and stability. It has been reported that CSB may act in the suppression of ROS-induced DNA breaks at telomeres in an R-loop-dependent mechanism[113]. Indeed, Tan et al. (2020) demonstrated that cells deficient in CSB experienced greater telomere aberrations than WT cells, particularly upon ROS-induced oxidative damage. During replication stress, CSB has also been reported to recruit HR repair proteins to telomeres to prevent fork collapse and telomere fragility[114]. Additionally, Batenburg et al. (2012) demonstrated a role for CSB in maintaining telomere length and stability in a TRF2-dependent manner, where CSB was required for the suppression of telomere doublets and telomere dysfunction-induced foci (TIFs), both of which result in telomere instability[115]. Together, these results indicated that CSB plays a role in preventing the onset of aging through various mechanisms, although the roles of CSB in these pathways remain unclear.

2.3. Mismatch Repair (MMR)

The MMR pathway is primarily responsible for the recognition and repair of nucleotide misincorporations generated during DNA synthesis, resulting in either base-base mismatches or small insertion or deletion loops[116]. The functions of and steps involved in MMR have been reviewed extensively[116-118]. Briefly, the misincorporated base is recognized by either the MutS Homologue 2 (MSH2)/MutS Homologue 6 (MSH6) complex (MutSα) or the MSH2/MutS Homologue 3 (MSH3) complex (MutSβ), depending on the mismatch or loop structure. These complexes then recruit other factors, including MutL Homologue 1 (MLH1), PMS1 Homologue 2 (PMS2), and Exonuclease 1 (EXO1) for the removal of both the mismatched base and/or loop and several bases downstream of the damage. The repair factors PCNA, RFC, RPA, Polδ/ε, and DNA ligase then work together to fill in the gap to complete the process. A number of factors can influence the rate of nucleotide misincorporation, including the fidelity of the DNA polymerase (Figure 3). Replication polymerases typically have a high fidelity, generating few errors; however, translesion polymerases often have lower fidelity, generating mismatches more frequently than replication polymerases[116]. Similar to the other DNA repair pathways, MMR is critical for maintaining genomic stability and erroneous MMR can have deleterious consequences, including the generation of a mutator phenotype[119, 120]. Cells are constantly undergoing DNA replication and without MMR, mismatches generated by both high and low fidelity polymerases will accumulate, resulting in increased mutations, cellular senescence, and cell death. Defects in MMR have also been linked to several forms of cancer; in particular, MMR defects are frequently observed in patients with colorectal cancer (CRC)[121]. Obesity is recognized as a high-risk factor for the development of CRC[122]; however, the relationship between obesity and MMR in CRC is not well understood.

2.3.1. MutL Homologue 1 (MLH1)

MLH1 forms a heterodimer with PMS2 that functions primarily in the MMR pathway. This complex, known as MutLα, acts as a validation step, improving the recognition of the MutSα or MutSβ complexes to mismatched or looped DNA, respectively. MutLα also possesses cryptic endonuclease activity, which is thought to be required for the removal of the mismatched base via EXO1[116]. The lack of redundant binding partners for the formation of MutLα suggests that loss of MLH1 would hamper the MMR process, preventing the repair of mismatched or looped DNA[123]. Indeed, MLH1-deficient mice generated by Edelmann, et al. (1996) were replication-error positive (RER⁺) and demonstrated increased microsatellite instability and decreased MMR activity[124]. Additionally, both male and female MLH1-deficient mice were sterile due to meiotic disruption and apoptotic loss of germ cells. With an impaired MMR pathway, it was likely that the MLH1-deficient mouse germ cells accumulated too much DNA damage during meiosis, thereby resulting in apoptosis during the meiotic checkpoint. Further emphasizing the importance of MLH1 in the maintenance of genome integrity, Hegan et al. (2006) generated MLH1 knockout mice containing a recoverable mutation reporter and demonstrated that MLH1-deficient mice showed significantly higher mutation frequencies in skin and colon tissues than their WT counterparts[125].

Concurrent with genomic instability representing a hallmark of aging, several groups have reported associations between declining MMR activity and age[126-129]. Among the first to propose a relationship between MLH1 and age, Kim et al. (2006) analyzed and compared the coding region of MLH1 in individuals over 100-years old to a control group of ~50-years old and found that the centenarian population had an increased MLH1 haplotype frequency compared to the control group[127]. Although no potential mechanisms were proposed for this association, the outcome led to the idea that perhaps a stabilization of genome maintenance enzymes, particularly MMR enzymes, could enhance longevity. Likewise, a decrease in repair activity could lead to reduced longevity due to increased genomic instability. In support of this, Edelman et al. (1999) reported striking longevity differences between mice with MLH1 heterozygous genotypes and MLH1 null genotypes compared to WT littermates, with ~50% of the mice perishing at 18 months and 7 months, respectively[130]. Later, Kenyon et al. (2012) demonstrated decreased gene expression and protein levels of MLH1 that correlated with increased microsatellite instability events with age in hematopoietic stem cells[126]. The decrease in gene expression was attributed to epigenetic regulation, as the promoter region of MLH1 was increasingly methylated with age. This was an interesting finding, as it provided evidence for a proposed mechanism by which age-induced MMR deficiency could lead to increased genomic instability via transcriptional suppression of a specific DNA repair enzyme. While there is a foundation of evidence to support a role for MLH1 in aging, further studies are warranted, particularly in the context of animal models to investigate the underlying mechanisms of age-associated suppression of MLH1, and how this suppression could further exacerbate aging phenotypes.

The relationship between MLH1 and obesity remains unclear. Amongst the few published studies that have investigated this relationship, most have had a narrowed focus on the impact of lifestyle factors, such as diet, on MMR-deficient colon cancers. Campbell et al. (2009), for example, reported that a western diet pattern increased colon cancer risk in individuals with MLH1 polymorphisms[131]. Western diet patterns, characterized by increased consumption of high-fat and high-sugar foods, have previously been associated with increased risk for obesity and metabolic syndrome compared to traditional or Mediterranean diets[132, 133]. A randomized controlled trial reported an association of obesity and increased risk of colon cancer in patients with Lynch syndrome, a disorder that is caused by a defect in one or more of the MMR genes, typically MLH1, MSH2, and/or PMS2[118, 134]. Taking a more direct approach, Remely et al. (2017) reported a decreased expression of MLH1 and an increased level of CpG promoter methylation in mice fed high-fat diets compared to those on control diets[135]. Together, these data suggest underlying regulatory relationships between obesity and MLH1 status and prompt further investigations into how these two interact, particularly in the context of colon cancer development.

2.3.2. MutS Homologue 2 (MSH2)

MSH2 binds to either MSH6 or MSH3 to form the MutSα and MutSβ heterodimers, respectively, that can recognize mismatched or looped DNA and initiate the MMR mechanism. Like other proteins involved in MMR, defects in MSH2 can result in genomic instability and increased risk of several cancers. Similar to MLH1, MSH2 was found to be downregulated in aging hematopoietic stem cells[129]. Additionally, 12-month-old mice showed both increased hypermethylated MSH2 promoter regions and decreased MSH2 mRNA levels compared to 2-month-old mice, indicating a possible role for MSH2 in the aging process[136]. Further, mice deficient in MSH2 demonstrated increased mutation frequencies in spleen and colon tissues compared to their WT counterparts, similar to the observations seen in MLH1-deficient mice[125]. These results suggested that, like MLH1, deficiency in MSH2 may promote the aging process by an increased occurrence in microsatellite and genomic instability events. In support of this idea, Campbell et al. (2006) used MEFs from MSH2-deficient mice to study the possible roles of MMR, and specifically MSH2, on telomere maintenance and chromosome stability[137]. Indeed, a significant increase in chromosomal aberrations, including centrosome amplification, chromosome mis-segregation, and aneuploidy events were observed in MSH2-deficient MEFs compared to WT MEFs. Interestingly, MSH2 deficiency had only a slight impact on telomere capping defects, and no impact on telomere length or telomerase regulation. Thus, these results suggested that rather than having a direct effect on telomere maintenance, mutations in telomere sequences, and thus telomere function, may be due to the increased microsatellite instability events that occur due to defects in MMR.

As with MLH1, the relationship between MSH2 and obesity remains largely unknown. Most of the reported studies have focused on the impact of diet and gut microbiota on the development of CRC in MSH2-deficient systems. Dietary factors play a large role in the development of obesity and can lead to significant alterations in the microbiome and increased risk for the development of human diseases, as previously reviewed[138, 139]. Indeed, researchers found that the gut microbiome of mice genetically predisposed to CRC (i.e., MSH2-deficient mice with mutated APC) lead to polyp formation, hyperproliferation, and epithelial cell transformation, precursors for cancer development[140]. Interestingly, these events were significantly reduced when the mice were fed a low-carbohydrate diet, suggesting that the metabolites produced by the microbiome on diets reminiscent of western diet patterns were a driving force in the development of CRC[140]. While these experiments provide a solid foundation for understanding some of the mechanisms underlying the relationship between diet and cancer development, further studies are warranted to investigate the role(s) of MSH2 in obesity-related diseases and how these mechanisms can be utilized for the prevention and/or treatment of human disease.

2.4. Non-homologous End-joining (NHEJ)

Generated from both endogenous and exogenous sources, DSBs are among the most toxic and deleterious lesions that can occur in DNA. Mis-repair of these lesions can have severe consequences on the integrity of the genome and lead to a variety of mutations, including large deletions and chromosomal translocations[141]. Thus, it is imperative that DSBs are identified and repaired efficiently and in an error-free fashion. NHEJ is the primary mechanism for the repair of DSBs outside of S-phase. The NHEJ pathway and the proteins involved have been reviewed extensively[142-145]. Briefly, broken ends are recognized by the Ku heterodimer, composed of Ku70 and Ku80. Once bound, this heterodimer recruits most of the other components of the NHEJ pathway, including DNA-PKcs, X-ray cross-complementing protein 4 (XRCC4), DNA Ligase IV, XRCC4-like factor (XLF), and numerous other proteins required for generating compatible ends prior to ligation. Once compatible ends have been generated, XRCC4, DNA ligase IV, and XLF work together to seal the ends, effectively repairing the DSB (Figure 3). [142].

Often considered an error-prone process, studies have recently differentiated two primary mechanisms of repair with varying levels of fidelity; classical NHEJ (c-NHEJ), which is relatively error-free, and alternative NHEJ (A-NHEJ), which is generally more error-prone[146, 147]. Currently, it is unclear as to why one pathway is chosen over the other. During states of increased DNA damage or dysregulated DNA repair, such as with aging and/or obesity, it could be that the more mutagenic A-NHEJ pathway is able to compete with c-NHEJ and perpetuate the cycle of genetic instability. Several groups have observed that with age, the efficiency of NHEJ decreases, resulting in an accumulation of genomic instability[148-152]. Transgenic mice deficient in critical NHEJ proteins also demonstrated signs of premature aging, including proliferative defects, cellular senescence, and telomere shortening, all of which are prominent hallmarks of aging[153-156].

Reports demonstrating changes in the levels of NHEJ proteins with age are contradictory. Some groups have observed decreased Ku levels in human fibroblasts and lymphocytes with age and senescence[157, 158]. Other studies, however, reported that changes in Ku expression and activity were tissue-specific, with kidney, lung, testis, and liver displaying varying degrees of age-associated modulation[159]. In contrast, Li et al. (2016) analyzed the levels of several NHEJ proteins in 50 human eyelid fibroblasts from young and old doners and reported that XRCC4 and Ligase 4 proteins decreased with age, while Ku protein levels remained unchanged[152]. This observation of decreased XRCC4 expression with age has also been observed across several tissues in rats[160]. The cause for the discrepancies between these studies is unclear; however, it is evident that the modulation of NHEJ factors may be influenced by the aging process.

Interestingly, while both of the rodent studies mentioned above reported different proteins being affected by age, calorie restriction reduced the observed effects in both experiments, returning the expression and activity of the DNA repair factors to baseline levels[159, 160]. Ke et al. (2020) investigated this phenomenon further by placing NHEJ-reporter mice on short-term calorie restricted diets and analyzing NHEJ efficiency across multiple tissues[161]. Mice on calorie restricted diets had statistically significant increases in NHEJ repair efficiency in skin, lung, kidney, and brain, as well as increased protein levels of DNA-PK compared to mice on control diets. Although the diets were short-term, this study provided strong evidence that dietary intervention could aid in the prevention of an age-dependent decline in DNA repair and prompts further investigation into the impact of long-term calorie restriction on DNA repair efficiency.

Although there are few reported studies on calorie restriction and NHEJ, there are even fewer investigating the potential impact of obesity on NHEJ repair. Yu et al. (2018) reported that although mice on HFD demonstrated an accumulation of DSBs and apoptosis, NHEJ appeared unchanged[162]. It is important to note that this study used 53BP1 staining as an indicator of functional NHEJ instead of a transgenic-reporter mouse model. Given the effects of calorie restriction on NHEJ, and the accumulation of DSBs in mice on HFD, it would be worth investigating further for any potential relationships between obesity and NHEJ. The use of NHEJ-reporter mouse models or the development of functional NHEJ assays suitable for tissue extracts may provide a more unique and direct insight into the potential impact of obesity on NHEJ efficiency.

2.5. Homologous Recombination (HR)

In addition to NHEJ, HR is another primary mechanism for the repair of DSBs. However, unlike NHEJ, which can occur throughout the cell cycle, HR utilizes the sister chromatid as a template for error-free repair of broken DNA ends and thus only occurs during the S- and G2-phases of the cell cycle. Several recent publications have reviewed the current understanding of the complex and intricate processes underlying HR[163-166]. Briefly, the broken ends are recognized by a complex of proteins referred to as the MRN complex, including Meiotic recombination 11 (MRE11), Rad50, and Nijmegen breakage syndrome 1 (NBS1). Several additional proteins, including CtBP-interacting protein (CtIP), EXO1, Bloom Helicase-DNA replication ATP-dependent helicase/nuclease (BLM-DNA2) complex, and DNA repair protein Rad51 homologue 1 (Rad51) are then recruited for DNA end resection, strand invasion, and repair synthesis using the homologous sister chromatid as a template (Figure 3).

It is established that DNA damage, and in particular DSBs, increases with age. The extent to which this increase is due to an accumulation of DNA damage, decreased DNA repair, or a combination of both, remains unclear. Studies have demonstrated that, along with an increase in DSBs, the efficiency of HR repair decreases with age[152, 167]. Similar to NHEJ, there are conflicting reports in regard to the relative levels of HR repair proteins with age. Some groups have reported decreased levels of key HR proteins with age, including MRE11 and Rad51, in yeast and human lymphocytes[157, 168]. In contrast, Li et al. (2016) observed no changes in HR repair protein levels in aged human eyelid fibroblasts, and instead proposed that the impaired recruitment of Rad51 to DNA damage sites contributed to decreased HR efficiency[152]. Interestingly, in addition to decreased HR repair, another group utilizing Drosophila as a model of aging observed increased expression and recruitment of Rad51 to DNA damaged sites with age, opposite to the observations of the previously mentioned studies[167]. As overexpression of Rad51 had previously been reported to result in chromosomal aberrations and genetic instability, the authors proposed that excess Rad51 could potentially delay filament disassembly during repair synthesis and stall the repair of the break[169]. The contradictory reports in the levels of HR repair proteins during the aging process are likely attributed to both differences in the cell type and species utilized in the published studies.

Aside from alterations in protein levels, there is a growing body of evidence suggesting the involvement of the HR repair pathway in telomere maintenance. The MRN complex has been reported to negatively regulate access of TRF1 to telomere ends, allowing telomerase access to the telomere ends and promoting telomere lengthening[170]. Depletion of Rad51 and BRCA2 resulted in telomere attrition, fragility, and telomere-associated chromosomal abnormalities in various models, including fungus, MEFs, and human cells[171-173]. Rad51-depleted human cells showed increased 53BP1 foci formation and telomeric fragments after CRISPR-Cas9-induced DSBs, indicating decreased DNA repair and providing further evidence for the involvement of HR in telomeric DSB repair[174]. Mice deficient in Rad54, a Rad51-interacting protein during HR, demonstrated telomere shortening and chromosomal aberrations, suggesting a role in telomere protection[175]. Most recently, Feretzaki et al. (2020) reported that Rad51 facilitated the recruitment of TERRA, a long non-coding RNA reported to be involved in telomere maintenance, to telomere ends via R-loop formation[176, 177]. Together, it is evident that HR repair proteins play protective roles in preventing telomere fragility and instability, prominent hallmarks of aging.

While few in number, multiple reports have demonstrated increased DSBs, as assessed by γ-H2AX markers, during obesity[11, 162, 178, 179]. Moreover, Yu et al. (2018) reported evidence of decreased HR repair via fewer Rad51-positive foci in mice fed HFDs compared to control chow diets[162]. Although still requiring additional mechanistic evidence, these studies provide a foundation for future experiments to further investigate the relationship of obesity and DSB repair, and in particular how HR repair is impacted under conditions of obesity.

3. DNA Replication

DNA replication is among the most studied of the DNA metabolic processes and must occur before cell division can be completed. With trillions of cells dividing each day, it is imperative that the fidelity of DNA replication be preserved. Countless proteins are involved in ensuring accurate and efficient replication of the genome, and their dysregulation can result in disastrous consequences. Further, endogenous and exogenous factors can result in the stalling of the replication process. Unresolved replication stress can result in DNA damage, genomic instability, and cell death. In this section, we will discuss a few of the many proteins that are involved in maintaining DNA replication fidelity and their roles in aging and obesity, as well as how aging and obesity can modulate the functions of these proteins (see Table 1).

3.1. Werner Syndrome ATP-dependent Helicase (WRN)

A member of the RecQ family of helicases, WRN is a multifunctional protein involved in numerous DNA metabolic processes. Werner syndrome, the result of mutations in WRN, is characterized by premature aging and an increased risk of cancer[180, 181]. Additionally, patients with Werner syndrome are also at an increased risk for early onset T2DM, cataracts, and skin atrophy[180, 182]. Human cells depleted of WRN and primary cells from WRN patients displayed increased mitochondrial ROS, decreased NAD+ levels, and loss of mitochondrial morphology, indicating severe mitochondrial dysfunction[183]. Restoration of NAD+ to normal levels reversed these phenotypes and inhibited the accelerated aging observed in the cells[183]. Dysregulation of NAD+ levels and altered mitochondrial function are prominent in metabolic and premature aging disorders[184]. These findings suggested that mitochondrial dysfunction may be a significant contributor to the accelerated aging phenotypes observed in patients with Werner syndrome.

Helicase and exonuclease domains allow for the involvement of WRN in several DNA processes, including DNA replication, repair, and telomere maintenance[185], suggesting a primary function for WRN in maintaining genomic integrity. In support of this, primary fibroblasts from Werner syndrome donors demonstrated proliferative defects and early onset of cellular senescence compared to WT controls[186]. These results were recapitulated in fibroblasts depleted of WRN, showing a direct relationship between WRN function and the observed phenotypes[186]. Further, WRN was reported to be involved in the resolution of multiple alternative DNA structures, as depletion of WRN in human cells lead to increased DNA structure-induced genetic instability[187, 188]. There are many causes of early onset of cellular senescence, including DNA damage and telomeric dysfunction. Cells lacking WRN helicase activity showed defective telomere replication and deletions in telomeric regions[189]. Additional studies have reported roles for WRN in the repair of telomeric DNA damage, as cells depleted of WRN were more susceptible to telomeric DNA damage, telomere fusions, and telomere dysfunction[190-192]. Interestingly, a study by Shamanna et al. (2016) reported that WRN deficiency decreased NHEJ efficiency and promoted the more mutagenic A-NHEJ pathway, resulting in increased DNA end resections and telomere fusions in WRN-deficient MEFs[193]. These results suggested that WRN may play a role in determining pathway choice in DSB repair, and that increased A-NHEJ repair in WRN-deficient cells may be a contributing factor to the increased genomic instability and telomere dysfunction phenotypes observed in this progeroid disease.

Compared to WT control mice, WRN-deficient mice placed on HFD demonstrated increased weight gain, hyperinsulinemia, and insulin resistance, all signs of obesity and metabolic syndrome[194]. Although WT mice fed a HFD also showed similar phenotypes, they were not as severe as those observed in WRN-deficient mice. Another study reported by Massip et al. (2006) found that mice with a mutation in the helicase domain of the WRN protein had increased weight gain, hyperinsulinemia, triglyceride deposits, and cholesterol levels compared to WT mice[195]. The results from this study were particularly interesting, as they demonstrated that even in the absence of a HFD, mice with mutated WRN still developed clear signs of obesity and metabolic disorder. Interestingly, most patients with Werner syndrome have dysregulated insulin signaling, which is thought to be the main contributing factor for the early onset T2DM observed in these patients[194-196]. Further, the increase in insulin levels in mice deficient in WRN appeared to be slightly exacerbated with age, indicating that age-related alterations in metabolic signaling may also occur in the absence of functional WRN protein[197].

3.2. Bloom Syndrome (BLM) Protein

BLM is another RecQ helicase involved in maintaining genome stability and integrity. Deficiency in BLM leads to the developmental disorder known as Bloom syndrome, which is characterized by stunted growth, insulin resistance, immune deficiencies, and sensitivity to sunlight[198, 199]. As characteristics of Bloom syndrome are not consistent with other progeroid syndromes (e.g., graying hair, cataracts, skin atrophy, etc.), it is not currently classified as a progeroid disorder[200]. However, individuals with Bloom syndrome are far more susceptible to developing cancer, widely accepted as a disease of aging, and often die young due to health complications[198-200].

Multiple groups have demonstrated that knockdown of BLM in human primary fibroblast cells resulted in early onset of cellular senescence, a notable hallmark of aging and common phenotype of aging disorders[201, 202]. In vitro studies have demonstrated that BLM interacted with telomere-associated proteins such as TRF2, which stimulated its helicase activities[203]. BLM was also found to localize to a subset of telomeres in human cells, although telomere repeat lengths in Bloom syndrome cells were not significantly different than normal cells, suggesting its involvement may be more related to a surveillance mechanism than length maintenance[204]. However, a study by Barefield et al. (2012) demonstrated BLM association to telomeres in G2/M phases, and that lack of BLM resulted in chromosomal abnormalities, including chromosome fusions[205]. The association of BLM during late cell cycle phases suggests that BLM may be involved in post-replicative maintenance of telomeres. BLM has been previously reported to be involved in the HR repair pathway, which is only active during S/G2 phases when a sister chromatid is available, suggesting that the involvement of BLM in telomere homeostasis may be largely due to its involvement in DSB repair[206-208]. Further, BLM was reported to be involved in the resolution of alternative DNA structures as well as replication fork restart after replication stress, preventing DSB formation and preserving genome integrity[209, 210]. Together, these studies provide clear evidence that BLM is critical for maintaining genome stability.

Compared to WRN, there are fewer studies published on the roles of BLM in obesity and metabolic syndrome. In fact, most children with Bloom syndrome are often underweight and display a lack of appetite, although this is typically not carried over into adulthood with some adults developing obesity[199, 211]. Similar to WRN, many individuals with Bloom syndrome develop T2DM due to insulin resistance, although the mechanisms underlying the involvement of BLM in insulin signaling are currently unclear, and warrant further investigation[198].

3.3. Flap Endonuclease 1 (FEN1)

FEN1 is a multifunctional, structure-specific metallonuclease involved in several DNA metabolic processes, including lagging strand synthesis in DNA replication and long-patch BER. Its primary function is to resolve intermediate DNA structures to prevent genomic instability[17, 212]. Cells depleted of FEN1 often display triplet repeat instability, DNA-structure-induced mutagenesis, replication defects, and decreased cell survival[19, 212-214]. Further, mice with complete knockout of FEN1 are embryonically lethal and mice haploinsufficient for FEN1 are more susceptible to developing cancer, underscoring its critical roles in development and genome stability[215]. While there are limited published reports investigating FEN1 and obesity, a study published by Larsen et al. (2008) reported that mice with a mutated FEN1 nuclease domain developed obesity by 9 months of age compared to WT control mice[216]. The absence of alterations in insulin or glucose levels suggested that the development of obesity was independent of insulin resistance or diabetes. Currently, studies are lacking on the potential involvement of FEN1 in the development of obesity, or how the obese state could regulate FEN1 activity.

As mentioned throughout this review, prominent hallmarks of aging include telomere fragility and genomic instability, and many studies have demonstrated a role for FEN1 in these events. For example, a study by Saharia et al. (2008) demonstrated that depletion of FEN1 in human cells resulted in increased DNA damage at telomere sequences as well as telomere loss[217]. Further investigation revealed that telomere loss was observed in telomeres that underwent lagging strand synthesis, suggesting that telomere dysfunction in FEN1-depleted cells was due to replication stress. The WRN protein has been previously shown to associate with and stimulate the cleavage activity of FEN1 at telomeres[217, 218]. WRN has also been reported to be essential for replication fork restart, as well as forming a complex with FEN1 during replication fork stalling to process branch structures at the fork[219, 220]. Saharia et al. (2010) demonstrated that FEN1-depleted HeLa cells had less efficient replication fork restart after hydroxyurea treatment than WT control cells[221]. Therefore, it is likely that FEN1 is involved in the initiation of stalled replication forks during telomere replication along with WRN to prevent telomere dysfunction and instability.

Although its primary roles involve maintaining DNA replication integrity, FEN1 is also likely to be recruited to telomeres in response to DNA damage via roles in DNA repair pathways. Indeed, a study observing interactions of telomere binding proteins with DNA repair proteins revealed that TRF2, a telomere capping protein, associated with both FEN1 and Polβ[222]. Further, Sun et al. (2015) found that FEN1, along with Polβ and NTH1, was recruited to telomeres following site-specific DNA damage[223]. These data suggest a potential role for FEN1 in the resolution of DNA damage at telomeres via its involvement in the long-patch BER sub-pathway.

3.4. Twinkle Helicase

Twinkle is a mitochondrial DNA helicase critical for the replication of mitochondrial DNA. It functions primarily to unwind mitochondrial DNA at the replication fork and is thought to be the sole essential replicative helicase for mitochondrial DNA replication[224, 225]. Mutations in Twinkle result in mitochondrial diseases, such as progressive external ophthalmoplegia (PEO), mitochondrial DNA depletion syndromes, and Perrault syndrome[226-229]. Mitochondrial diseases are characterized by decreased mitochondrial respiration and mitochondrial dysfunction, common features observed in both aging and obese states. Some overlap exists in the phenotypes observed from Twinkle mutations and mutations in polymerase-γ (POLG). This is not unexpected, as Twinkle works in concert with POLG and the mitochondrial single-stranded DNA binding protein (mtSSB) to replicate mitochondrial DNA[226, 230]. Therefore, it is not surprising that mutations in key replisome factors would result in overlapping diseases, as they play crucial roles in the maintenance of mitochondrial DNA.

Along with a decline in mitochondrial respiration, aging mice also showed marked decreases in Twinkle protein levels and mitochondrial DNA copy number[231]. Additionally, a study analyzing mitochondrial DNA copy number in peripheral blood cells from over 1,000 patients demonstrated decreased copy number with age[232]. Mice overexpressing the Twinkle helicase have been reported to have increased mitochondrial DNA copy number[224]. Based on these findings, Foote et al. (2018) set out to determine if restoration of mitochondrial DNA copy number by overexpression of Twinkle could rescue the mitochondrial dysfunction observed in aging mice. Indeed, mice overexpressing Twinkle displayed significantly increased mitochondrial copy numbers and respiration rates compared to age-matched WT mice[231]. Further, mice overexpressing Twinkle appeared to have delayed vascular aging compared to age-matched WT mice, as demonstrated by a delay in the decline of arterial distensibility and arterial compliance. Together, these results highlight an interesting mechanism by which aging phenotypes such as mitochondrial dysfunction and vascular aging may be ameliorated by restoring mitochondrial DNA copy number and replication fidelity.

4. Concluding Remarks

DNA repair and replication proteins play crucial roles in the maintenance of genome integrity, and as such, deficiencies or dysregulation of these proteins can result in aberrant DNA repair, genetic instability, and the development of several diseases. Here, we have reviewed the existing literature investigating the roles and functions of various repair and replication proteins in aging and obesity, and how these proteins may be involved in the mechanisms underlying aging- and obesity-related disorders.

Some of these proteins, such as the OGG1 and NEIL proteins, have shown obvious roles for preventing common hallmarks of both aging and obesity, including the regulation of mitochondrial function, modulation of energy homeostasis, and prevention of genomic instability. Others, such as XP proteins, CS proteins, and the RecQ helicases WRN and BLM, have demonstrated roles primarily in the prevention of aging phenotypes, in which loss of these proteins leads to various progeroid syndromes. Nevertheless, the studies and findings discussed in this review provide clear and strong support for links between DNA repair fidelity, aging, and obesity and provide a solid foundation for further studies to investigate the underlying mechanisms involved.

Highlights.

• Aging and obesity can lead to the dysregulation of DNA repair and replication proteins and/or pathways

• Aging and obesity can result in increased DNA damage and decreased DNA repair, resulting in an accumulation of mutations and genomic instability

• Deficiencies and/or the dysregulation of DNA repair and replication proteins can result in aberrant DNA repair, genomic instability, and the development of several diseases

• Mutations in genes responsible for maintaining genome integrity can exacerbate aging- and obesity-related phenotypes and disorders