Peptide Hormone Regulation of DNA Damage Responses

Abstract

DNA damage response (DDR) and DNA repair pathways determine neoplastic cell transformation and therapeutic responses, as well as the aging process. Altered DDR functioning results in accumulation of unrepaired DNA damage, increased frequency of tumorigenic mutations, and premature aging. Recent evidence suggests that polypeptide hormones play a role in modulating DDR and DNA damage repair, while DNA damage accumulation may also affect hormonal status. We review the available reports elucidating involvement of insulin-like growth factor 1 (IGF1), growth hormone (GH), α-melanocyte stimulating hormone (αMSH), and gonadotropin-releasing hormone (GnRH)/gonadotropins in DDR and DNA repair as well as the current understanding of pathways enabling these actions. We discuss effects of DNA damage pathway mutations, including Fanconi anemia, on endocrine function and consider mechanisms underlying these phenotypes. (Endocrine Reviews 41: 1 – 19, 2020)

Graphical Abstract

Graphical Abstract.

Essential Points

• Efficient DDR is important for maintaining genome integrity and cell viability

• Peptide hormones act to mediate DDR and DNA repair efficacy and fidelity

• Hormonal effects on DNA damage may be cell-type or tissue-specific

• Some polypeptide hormones elicit opposing actions on DDR and DNA repair in either malignant or in nontransformed cells

• Compromised DNA damage response and accumulated DNA damage are linked to endocrine abnormalities, including somatotroph axis suppression

• As peptide hormones regulate DNA repair efficacy, hormonal status may play an important role in resistance to DNA-damaging therapy

The DNA damage response (DDR), a complex, highly regulated process that detects and repairs DNA damage, ensures low rates of spontaneous mutations during mitosis; compromised DNA repair results in accumulated mutations (1, 2). In malignant tumors, DDR is frequently altered, resulting in an inability to limit DNA damage, thereby enabling a permissive state for potentially oncogenic mutations. DDR genes acting as “caretakers” of the genome (3) detect DNA damage and either activate repair machinery or directly repair DNA damage (1, 4, 5). As defective DDR affects tumor predisposition and alters sensitivity of tumors to chemotherapy (6), defects in maintaining genome integrity and in repair mechanisms are selectively advantageous for neoplastic progression (1).

Although DDR processes in response to DNA damage are well described, little is known about cellular signaling mechanisms that modulate effectiveness of DNA repair. Recent evidence suggests a role for hormones and their respective receptors in DDR regulation. Steroid hormone effects in DDR were recently reviewed in detail by Schiewer and Knudsen (7), and emerging evidence now also suggests that peptide hormones are functionally linked to DDR regulation (Fig. 1A). We review available results on effects of insulin-like growth factor 1 (IGF1), growth hormone (GH), α-melanocyte stimulating hormone (αMSH), and gonadotropins on DDR and DNA repair, discuss effects of DNA damage accumulation on hormonal status, and consider proposed mechanisms underlying DDR-hormone communications.

Figure 1.

Simplified representation of mechanisms of DDR and DNA repair. Shown are types of DNA damage and repair known to be affected by GH, IGF1, and αMSH (see Tables 1-3). (A) Replication stress, oxidative damage, UV irradiation, chemotherapy, and radiation therapy lead to different types of DNA damage, including double-strand breaks (DSBs) that can lead to chromosomal aberration if left unrepaired. (B) DDR is facilitated by activation of phosphatidylinositol 3-kinase-related kinase family members ATM and ATR, DNA-PKcs, and PARP family members, which phosphorylate the tumor suppressor p53 and checkpoint kinases Chk1 and Chk2 to block cell-cycle progression and slow propagation of damaged cells. Activation of the DDR pathway promotes activation of DNA repair mechanisms driven by specific kinases, which enables restored cell-cycle progression following DNA repair. Abbreviations: ATM, ataxia-telangiectasia-mutated; ATR, ataxia telangiectasia and Rad3-related; BER, base excision repair; BRCA, breast cancer; CPD, cyclobutane pyrimidine dimers; DNA-PKcs, DNA-dependent protein kinase, catalytic subunit; DSB, double-strand break; HR, homologous recombination; NER, nucleotide excision repair; NHEJ, nonhomologous end joining; OGG1, 8-oxoguanin-DNA glycosilase; 8-OHdG, 8-hydroxy-2’-deoxyguanosine; p53BP1, p53 binding protein-1; PARP, poly(ADP-ribose) polymerase; SSB, single-strand break; TERT, telomerase reverse transcriptase.

DNA Damage Response

DNA damage is induced by environmental agents or generated spontaneously during DNA metabolism (4). DNA damage occurs as a result of deoxynucleotide misincorporation as well as interconversion or oxidation of DNA bases; DNA breaks are also generated by reactive oxygen species (ROS) (4, 8).

Exogenous DNA damage resulting from ultraviolet (UV) light or ionizing radiation or medical radiotherapy generates single- (SSBs) or double strand breaks (DSBs). Chemical agents used in chemotherapy produce SSBs and DSBs as well as intrastrand or interstrand crosslinks (ICLs); adducts and oxidative damage are observed with cigarette smoke (4) (Fig. 1A).

The DDR pathway senses DNA damage and enables a controlled protective response by blocking cell-cycle progression, activating repair mechanisms, and/or inducing senescence or apoptosis. Cell-surface-bound and soluble molecules also induce DDR activation in healthy bystander cells (9).

DDR is facilitated by multiple pathways including phosphatidylinositol 3-kinase-related kinase family members, specifically ataxia-telangiectasia-mutated (ATM), ataxia telangiectasia and Rad3-related (ATR), and DNA-dependent protein kinase, catalytic subunit (DNA-PKcs), as well as poly(ADP-ribose) polymerase (PARP) family members (4, 10). ATM and DNA-PKcs are mostly activated by DSBs, while ATR is activated by SSBs; PARP is activated by both types of DNA lesions (11).

Upon sensing DNA damage, activated ATM, ATR, and DNA-PKcs phosphorylate and stabilize the tumor suppressor p53 to block cell-cycle progression and slow propagation of damaged cells. ATR and ATM also phosphorylate the checkpoint kinases Chk1 and Chk2, respectively, which phosphorylate p53 and arrest cell proliferation (12, 13) (Fig. 1B).

Subsequent responses to DNA lesions are regulated by specific DDR sub-pathways. For example, bulky lesions induced by UV light trigger ATR, phosphorylating Chk1 and p53 to block replication (14). DSB processing is more complex: ATM is autophosphorylated and activated by several molecular events, including binding of the Mre11-Rad50-Nbs1 (MRN) complex (15-17) as well as changes in chromatin structure (18, 19). Chk2, p53, and other DDR-associated kinases are phosphorylated, and histone H2 variant X (H2AX) is induced. The H2AX-dependent signaling cascade is critical to recruitment of DNA repair proteins (20).

Limited data are available characterizing regulation of DDR efficacy. Autophagy, serving as a cytoprotector (21), is activated in response to DNA damage (22-24). Starvation is one of the most well-described pathways activated during autophagy. The starvation-induced mammalian target of rapamycin (mTOR) pathway negatively regulates autophagy. ATM is activated by exposure to DNA-damaging agents, repressing mTOR and triggering autophagy (25). Concurrently, ROS-induced DNA damage activates PARP1, suppressing mTOR and also inducing autophagy (26). These protective mechanisms prevent DNA damage–induced apoptosis and may play a role in cellular chemoresistance.

MicroRNAs (miRNAs), a class of small regulatory RNAs, reduce repair capacity and enhance sensitivity to DNA damaging agents (27). For example, miR-421 attenuates ATM expression leading to altered cell cycle checkpoints, while miR-24 overexpression downregulates H2AX, and both miRNAs enhance cellular radiosensitivity (28).

Peptide hormones may represent another means for affecting DDR. Here, we discuss the different ways they modify DDR pathway efficiency depending on cell and tissue type.

p53 is a key molecule that determines cell fate—whether the cell will undergo DNA repair, apoptosis, or enter senescence. DNA damage kinases phosphorylate p53, initiating G1 cell cycle arrest by activating CDK inhibitor Cdkn1a (p21) and providing a short time frame for DNA damage repair before subsequent proliferation resumes (Fig. 1B). When DNA damage is too extensive to repair, p53 is upregulated and BAX and PUMA are increased to induce apoptosis (29). Alternatively, p53-activated p21 may induce premature cell cycle arrest, i.e., senescence, which results in reduced DDR efficiency (30). Senescent cells are metabolically active, but in a state of proliferative arrest, and express multiple proteins comprising the senescence-associated secretory phenotype (SASP) (31-33). SASP composition varies depending on the cellular context, but typically consists of cytokines and chemokines (e.g., interleukin-6, interleukin-8, Groα, Groβ), matrix metallopeptidases, and growth factors, including IGF1 and IGF binding proteins (32). These locally secreted proteins may exert cell-specific effects on proliferation and DDR of neighboring cells. We showed that the SASP also includes GH, secreted by senescent human mammary adenocarcinoma and colon adenocarcinoma cells (34). The impact of IGF1 and GH pathways in the DDR is discussed below.

DNA damage can be directly measured experimentally by single-cell electrophoresis using the Comet assay (35) which detects both DSBs and SSBs; demonstration of free radical-induced oxidative lesions 8-hydroxy-2’-deoxyguanosine (8-OHdG), a reliable marker of oxidative DNA damage (36); expression of phosphorylated H2AX (37); or by assessment of chromosome aberration (38).

DNA Repair

Repair mechanisms are uniquely specific to types of DNA damage. For example, mispaired DNA bases are repaired with corrected bases, while nucleotide excision repair (NER) removes UV light-induced photoproducts, bulky chemical adducts, and intrastrand DNA crosslinks (8, 39). ICLs are detected and removed by the Fanconi anemia (FA)/BRCA pathway, and ICL processing results in adducts and DSBs, which are then repaired.

Subtle DNA changes include oxidative lesions, alkylation products, and SSBs. SSBs are common, arising at a frequency of tens of thousands per cell per day from direct effects of intracellular metabolites and ROS, or indirectly via enzymatic cleavage of the phosphodiester backbone. These breaks are repaired by base excision repair (BER), whereby damaged bases are removed from the double helix and the excised damaged DNA backbone is replaced with correctly synthesized DNA (4, 40, 41).

DSBs are the most lethal form of DNA damage and can lead to chromosomal aberrations and cellular transformation if left unrepaired. DSBs are repaired either by nonhomologous end joining (NHEJ) or homologous recombination (HR). NHEJ, a rapid and yet error-prone mechanism, reassembles broken DNA ends in the presence of DNA-PKcs. By contrast, HR, a high-fidelity repair mechanism initiated by ATM activation (4, 42), acts mainly in S and G2 to repair DNA gaps, DSBs, and ICLs, and restores original DNA sequences at the site of damage by resecting sequences around the DSB and using the homologous sister chromatid DNA sequence as a template for new DNA synthesis. Proteins encoded by BRCA1, BRCA2, RAD51, and PALB2 are required to mediate HR (43, 44).

p53 orchestrates several DDR mechanisms, including NER, BER, NHEJ, and HR (29, 45-47).

Peptide Hormone Regulation of DDR

Most available data on involvement of hormonal mechanisms in DDR regulation are derived from in vitro studies, which may be limited by supraphysiological oxygen levels as well as high medium glucose and hormone levels, both of which affect cell metabolism and, potentially, DDR. Many in vitro studies are also performed in malignant cells harboring signaling pathway mutations that may be involved in DDR and DNA repair. Additionally, discrepancies may be seen between in vitro and in vivo studies due to drug bioavailability and turnover, absence of plasma proteins, access to receptors, and timing of in vitro analysis. Thus, the in vitro experiments reviewed here serve as a starting point for understanding the complex relationships between hormonal status and DNA damage. We describe the most significant results obtained from in vitro studies as well as available data translating or extrapolating these mechanisms in vivo.

IGF1/IGF1 receptor (IGF1R) signaling

Table 1 lists in vitro and in vivo studies related to the effect of IGF1/IGF1R on DNA damage and repair (Figure 2).

Table 1.

Effect of IGF1/IGFR Signaling on DNA Damage and Repair

Author, Year (Reference)	Type of Study	Cells or Tissue	IGF1/IGF1R Manipulation	Type of DNA Damaging Treatment	Effects on DNA Damage and Repair	Mechanisms
Trojanek et al., 2003 (48)	In vitro	3T3-like fibroblasts from mouse embryos	IGF1 treatment	Cisplatin	↑ HR	RAD51 recruitment to DNA damage site
Yang et al., 2005 (49)	In vitro	Normal or transformed human mesangial cells	IGF1 treatment	Hyperglycemia-induced ROS	↑ HR	RAD51 recruitment to DNA damage site
Meyer et al., 2017 (50)	In vivo	Murine salivary gland cells	IGF1 treatment	Head and neck irradiation	↑ DNA repair	↑ Sirt1
Hèron-Milhavet et al., 2001 (52)	In vitro	Murine NWTb3 cells	IGF1R overexpression	4-NQO	↑ DNA repair
Loesch et al., 2016 (53)	In vitro Ex vivo explants In vivo	Primary human keratinocytes, N-TERT keratinocytes	IGF1R inactivation small molecule inhibitor	UVB irradiation	↓ NER	↓ XPC, ERCC3, ERCC4
Liu et al., 2018 (54)	In vitro	Human lung squamous carcinoma cells	IGF1R shRNAi	X-ray irradiation		↓ ATM ↓ H2AX ↓ p53 binding protein
Macaulay et al., 2001 (55)	In vitro	Murine melanoma cells	IGF1 antisense	γ Irradiation		↓ ATM kinase activity
Turney et al., 2012 (56)	In vitro	Human prostate DU145, PC3 cells	IGF1R siRNA	γ Irradiation	↓ HR
Rochester et al., 2005 (57)	In vitro	Human prostate DU145, PC3 cells	IGF1R siRNA	Mitoxantrone, etoposide, nitrogen mustard, γ irradiation	Enhanced sensitivity to DNA damage
Chitnis et al., 2014 (58)	In vitro	Human prostate DU145, PC3, 22Rv1 cells	IGF1R inactivating small molecule inhibitor	Cesium irradiation	↑ Nuclear fragmentation
		Human embryonic kidney (HEK293) cells			↓ HR ↓ NHEJ
Kemp et al., 2017 (59)	In vitro Ex vivo explants	Cultured human keratinocytes and skin explants	IGF1R inactivating small molecule inhibitor, IGF1 withdrawal	UVB irradiation		↓ Chk1 phosphorylation by ATR ↓ Protein A ↓ DNA replication
Cianfarani et al., 1998 (60)	In vitro	Human peripheral lymphocytes	IGF1 treatment	Bleomycin	↑ Chromosome aberrations
Fernandez et al., 2015 (61)	In vitro	Primary human keratinocytes and dermal fibroblasts	IGF1 treatment	UVB irradiation	↑ Removal of cyclobutene pyrimidine dimers (CPD)	↑ Chk1 ↑ pATR ↑ H2AX

Abbreviations: 4-NQO, 4-nitroquinoline 1-oxide; HR, homologous recombination; IGF1R, insulin-like growth factor 1 receptor; NER, nucleotide excision repair; NHEJ, nonhomologous end joining; p, phosphorylated; ROS, reactive oxygen species; siRNA, small-interfering RNA.

Figure 2.

IGF1 effect on DDR and DNA repair. Shown in black are mechanisms known to be affected by IGF1 (seeTable 1). IGF1 increases phosphorylation and activation of key proteins in the DDR that block cell-cycle progression and slow propagation of damaged cells, and activate key DNA repair mechanisms that enable restored cell-cycle progression following DNA repair (seeFig. 1B). Abbreviations: ATM, ataxia-telangiectasia-mutated; ATR, ataxia telangiectasia and Rad3-related; BER, base excision repair; BRCA, breast cancer; DNA-PKcs, DNA-dependent protein kinase, catalytic subunit; HR, homologous recombination; NER, nucleotide excision repair; NHEJ, nonhomologous end joining; OGG1, 8-oxoguanin-DNA glycosilase; p53BP1, p53 binding protein-1; PARP, poly(ADP-ribose) polymerase; TERT, telomerase reverse transcriptase.

In vitro studies.

IGF1 protects against cisplatin-induced DSBs in murine embryonic fibroblasts by enhancing HR proportional to IGF1R abundance. Specifically, IGF1 facilitates recruitment of RAD51 to DSBs to enhance HR repair (48). Normal and SV40-transformed human mesangial cells subjected to hyperglycemia-induced genotoxic stress exhibit both SSBs and DSBs; IGF1 treatment increased HR but not NHEJ, and enhancement of HR correlated with RAD51 translocation of foci of DNA damage (49). IGF1 may also induce expression of sirtuin 1 (50), implicated in DSB DNA repair (51).

UVB irradiation produces cyclobutane-pyrimidine dimers, photoproducts, and DNA strand breaks. IGF1 treatment of human keratinocytes and dermal fibroblasts irradiated with UVB resulted in increased expression of DDR proteins including phospho-ATR, Chk1, and H2AX concurrently with accelerated UVB-induced cyclobutene pyrimidine dimer removal (61). An opposite effect for IGF1 was demonstrated in a study of normal human lymphocytes treated with bleomycin, in which the addition of IGF1 dose-dependently increased the rate of chromosomal aberrations (60). Ligand activation of IGF1R in UV-damaged NIH-3T3 derived murine cell line NWTb3 overexpressing wild-type (WT) human IGF1R resulted in enhanced DNA damage repair and increased survival (52).

Confirmatory effects of IGF/IGFR on DNA repair were observed in 2 recent studies of primary normal human keratinocytes subjected to UVB. Inactivation of IGF1R using the specific IGF1R inhibitor AG538 inhibited NER-associated DNA repair genes, including XPC, ERCC3, and ERCC4 (53). When IGF1R signaling was abrogated, ATR phosphorylation of Chk1, which arrests cell proliferation, was attenuated (59, 62).

Radiation of primary murine glioma stem cells increased IGF1/IGF1R expression, which promoted Akt-dependent survival, thereby protecting cells from radiation damage. However, after treatment with an IGFR inhibitor, tumors formed from glioma stem cells showed increased radiosensitivity and decreased cell survival, indicating enhanced DNA damage or/and apoptosis (63). These results suggest that inactivation of IGF1R signaling increases sensitivity to DNA damaging agents.

Several studies have suggested potential mechanisms for IGF1 regulating DNA repair. When IGF1R was suppressed by antisense oligonucleotides, murine melanoma cells failed to induce ATM kinase activity after irradiation (55), suggesting that IGF1R modulates ATM function. This may explain decreased DNA repair seen in IGFR-deficient cells (56) and, conversely, induced HR DNA repair with IGF1 stimulation (48, 49). A role for IGF1 signaling in DNA repair was also demonstrated by suppressing IGF1R with small interfering RNA (siRNA) in human prostate cancer cells. Enhanced sensitivity to DNA damaging agents such as mitoxantrone, etoposide, and ionizing radiation was observed, producing SSBs and DSBs, while no increased sensitivity to chemotherapeutic agents that kill cells without damaging DNA was seen (57).

IGF1R effects on DNA repair are cell-specific yet broadly targeted across different repair mechanisms. For example, IGF1R depletion with siRNA or mutation reduces DSB repair by HR in human prostate cancer DU145 and PC3 cells, sensitizing these cells to ionizing radiation, while IGF1R-depleted LNCaP human prostate cancer cells showed comparatively modest radiosensitization (48, 56). Furthermore, treatment of human prostate cancer cells with a small molecule IGFR1 inhibitor induced radiosensitization associated with attenuated NHEJ and HR and reduced cell survival, while DNA damage increased nuclear fragmentation, consistent with late triggering of apoptosis (58). Finally, suppression of IGF1R led to ATM kinase inactivation, central to HR function (55), as well as to suppressed DNA-PKc phosphorylation, key to NHEJ (58). Although IGF1R mutant human colon Caco-2 cells did not show delayed repair of radiation-induced DNA as assessed by γH2AX expression, BRCA2, involved in HR, was suppressed in these cells (64).

A direct link between DDR and the IGF1/IGF1R pathway was demonstrated by use of short-hairpin RNA to decrease IGFR1 expression in human lung squamous cell carcinoma, which led to increased sensitivity to irradiation and decreased expression of ATM, γH2AX, and p53 binding protein-1, all associated with suppressed DNA repair pathways (54). In human mammary carcinoma MCF7 cells, IGF1 also indirectly affects DNA repair by enhancing p53-dependent increase in p21 to mediate growth arrest (65).

Importantly, ATM itself regulates the IGF1R. People with the inherited disorder ataxia telangiectasia harbor a mutation that eliminates or inactivates ATM, displaying neurodegeneration, growth abnormalities, premature aging, and increased sensitivity to radiation (66). Although ATM induces IGF1R promoter activity and IGF1R transcription, primary fibroblasts obtained from patients with the mutated kinase have reduced IGF1R expression, which likely contributes to the radiosensitivity and growth abnormalities seen in these patients (67). Similar effects of DNA-damage-induced ATM on IGF1/IGF1R activation were observed in MCF7 cells and in mouse embryonic fibroblasts (68).

The evidence presented above suggests that the IGF1/IGFR pathway, likely acting mainly through ATM, supports DNA damage repair in both cancerous and nontransformed cells (Table 1). Abrogated IGF1R signaling appears to sensitize human prostate tumor cells to chemo- and radiotherapy, leading to a proposal for IGF1R inhibition in treatment of resistant tumors and/or as an adjuvant therapy enhancing effects of DNA damaging treatments (58).

In vivo studies.

In contrast to the consistent observed favorable effects of IGF1/IGF1R for DNA integrity in vitro, data from in vivo studies are less consistent. Confirming effects of IGF1 on DNA damage in vitro (49), IGF1 pretreatment of salivary glands in mice undergoing head and neck radiation led to increased expression of intranuclear Rad51 and more rapid resolution of DNA damage (50). Although few animal models with altered IGF1 levels have been examined for changes in DDR, available evidence suggests an effect for IGF1 and IGF1R on DNA repair mechanisms. Pappa-knockout mice lack pappalysin, a metalloproteinase that cleaves IGF1-bound IGF1 binding protein 4 (IGFBP-4); Pappa deletion decreases IGF1 bioavailability due to increased IGFBP4 abundance (69). Skin fibroblasts derived from Pappa-knockout mice exhibit diminished mTOR1 activity, increasing DNA damage repair proteins O-methylguanine-DNA methyltransferase (MGMT) and N-myc downstream regulated gene 1 (NDRG1) (70). Concordantly, mouse embryonic fibroblasts derived from long-lived hemizygous IGF1R knockout mice show resistance to oxidative stress (71), and individuals with exceptional longevity are enriched with a functional mutation in IGF1R that confers partial IGF1 resistance (72, 73). (This latter role for IGF1 in aging is discussed more fully below in the context of murine progeroid models with defective DNA repair mechanisms and decreased GH/IGF1 signaling.)

Due to the very limited number of in vivo studies, as well as the fact that 2 in vivo studies demonstrated opposite effects of IGF1 on DNA repair (50, 70), it is difficult to affirm whether in vivo effects of IGF1 differ from in vitro actions on DNA damage repair.

GH/GH receptor (GHR) signaling

Table 2 lists in vitro and in vivo studies related to the effect of GH/GHR on DNA damage and repair (Figure 3).

Table 2.

Effect of GH/GHR Signaling on DNA Damage and Repair

Author, Year (Reference)	Type of Study	Cells or Tissue	GH and/or GHR Manipulation	Type of DNA Damaging Treatment	Effects on DNA Damage and Repair	Mechanisms
Madrid et al., 2002 (76)	In vitro	Chinese hamster ovary (CHO-4) cells	GHR overexpression	Bleomycin, radiation	↑ DNA repair
Wu et al., 2014 (77)	In vitro	Human colon carcinoma HCT-8 cells	GH treatment	γ Irradiation	↓ DNA damage	↑ GADD45 ↑ APEN
Bougen et al., 2011 Bougen et al., 2012 (78, 79)	In vitro	Human mammary carcinoma MDA-MB-231, MDA-MB 435S, and T47D cells	GH overexpression	Mitomycin, irradiation	↓ DNA damage	↑ BRCA1 ↑ BRCA2 ↑ TERT
	In vitro	Human endometrial carcinoma RL95-2 cells	GHR antagonist treatment (component of pegvisomant)	Mitomycin, irradiation	↑ DNA damage
Hohla et al., 2009 (80)	In vitro	Human colon adenocarcinoma HCT116 cells	GHRH antagonist treatment		↑ DNA damage apoptosis	↑ Phospho-p53 ↑ PARP ↑ p21
Bayram et al., 2014 (81)	In vivo	Human peripheral lymphocytes from acromegaly patients	Excess endogenous GH production		↑ Chromosomal DNA damage	Oxidative stress
Fantini et al., 2017 (82)	In vivo	Human peripheral lymphocytes from healthy subjects	Supraphysiological GH treatment		↑ DNA damage
Elbialy et al., 2018 (83)	In vivo	Liver from transgenic zebrafish	Excess endogenous GH production		↑ DNA damage	↓ Multiple DNA repair genes from NER, HR and NHEJ pathways
Cianfarani et al., 1998 (60)	In vitro	Human peripheral lymphocytes	GH treatment	Bleomycin	↑ Chromosome aberration
Chesnokova et al., 2019 (84)	In vitro	Human normal colon cells	GH treatment	Etoposide	↑ DNA damage ↓ NHEJ ↓ HR	↓ pATM ↓ Phospho-p53 ↓ pChk2
	In vitro	Human normal colon cells	GHR antagonist treatment (pegvisomant)		↓ DNA damage	↑ pATM
	In vivo	Colon tissue from nude mice	GH-secreting xenograft		↑ DNA damage
	In vivo	Colon tissue from GHR knock-out mice			↓ DNA damage
Podlutsky et al., 2017 (85)	In vivo	Skin fibroblasts from 9-week-old GH-deficient Lewis dwarf rats			↑ DNA repair	↑ Gadd45a ↑ Xrcc5 ↑ Ercc6
	In vivo	Skin fibroblasts from 3-week-old GH-deficient Snell dwarf mice			↑ DNA repair
	In vivo	Skin fibroblasts from 9-week-old GH-deficient Lewis dwarf rats	GH treatment from 5 to 9 weeks of age	γ Irradiation	↓ DNA repair	↓ Gadd45b ↓ Mdm2 ↓ Bbc3 ↓ Xrcc5 ↓ Ercc6
Dominick et al., 2017 (70)	In vivo	Skin fibroblasts from GH-deficient Snell mice and GHR-knockout mice				↑ MGMT ↑ NDRG1 DNA repair proteins

Abbreviations: GH, growth hormone; GHR, growth hormone receptor; GHRH, growth hormone–releasing hormone; HR, homologous recombination; NHEJ, nonhomologous end joining; p, phosphorylated; PARP, poly(ADP-ribose) polymerase.

Figure 3.

GH effect on DDR and DNA repair in nontumorous and malignant cells. Shown in black are mechanisms known to be affected by GH (seeTable 2). (A) In nontumorous cells, GH decreases phosphorylation and activation of key proteins in the DDR that block cell-cycle progression and slow propagation of damaged cells, and also suppresses key DNA repair mechanisms that allow for restored cell-cycle progression following DNA repair (seeFig. 1B). (B) In malignant cells, GH decreases phosphorylation of key DDR and DNA repair proteins. Abbreviations: ATM, ataxia-telangiectasia-mutated; ATR, ataxia telangiectasia and Rad3-related; BER, base excision repair; BRCA, breast cancer; DNA-PKcs, DNA-dependent protein kinase, catalytic subunit; HR, homologous recombination; NER, nucleotide excision repair; NHEJ, nonhomologous end joining; OGG1, 8-oxoguanin-DNA glycosilase; p53BP1, p53 binding protein-1; PARP, poly(ADP-ribose) polymerase; TERT, telomerase reverse transcriptase.

In vitro studies.

A role for GH signaling in DDR was suggested with GH rescue of human endometrial adenocarcinoma AN3CA cells (74) as well as MCF7 cells from the DNA-damaging effects of doxorubicin by reducing apoptosis and increasing chemoresistance (75). More directly, GH-treated Chinese hamster ovary (CHO)-4 cells transfected with GHR showed increased DNA damage repair and protection from radiation- and bleomycin-induced cell death (76), while hGH-treated human colon adenocarcinoma HCT-8 cells showed reduced radiation-induced DNA damage as well as upregulated GADD45 and APEN proteins associated with DNA damage repair (77).

An abundance of local GH in breast and endometrial cancers and in hepatocellular carcinoma confers a more adverse mortality outcome (86-88). Building on these findings, autocrine hGH was shown to increase clonogenic survival and attenuate radiation-induced or mitomycin-induced DNA damage in human mammary MDA-MB-436S and T47D breast carcinoma cells, as well as in human RL95-2 endometrial carcinoma stable transfectants. Protective effects of GH were mediated by the JAK2 and c-Src family. Quantitative PCR analysis showed that forced expression of hGH in MDA-MB-436S cells induced DNA damage repair genes, including BRCA1, BRCA2, and TERT, while blocking GHR in RL95-2 cells increased sensitivity to DNA-damaging therapy (78, 79).

The role of growth hormone–releasing hormone (GHRH) in DDR has also been explored. When triple-negative human breast cancer MDA231 cells were treated with the competitive GHRH antagonist [N-acetyl-Tyr1, D-Arg2] fragment 1-29Amide, DNA laddering and apoptosis were increased (89). Similarly, treatment of HCT116 human colon cancer cells with the GHRH antagonist JMR-132 caused SSBs and DSBs as evidenced by Comet assay as well as by loss of mitochondrial membrane potential, increased p53 phosphorylation, and induced p21 and pro-apoptotic factors Bax and PARP, with subsequent cell cycle arrest (80).

These results indicate that GHRH/GH signaling supports DNA damage repair in cancer cells. Importantly, however, the opposite effect is observed in nontransformed cells and tissues, where GH appears to induce DNA damage. We showed that hGH increased DNA damage as assessed by Comet assay in nontransformed human colon cells, mammary epithelial cells, and 3-dimensional human intestinal organoids derived from induced pluripotent stem cells (84). In these settings, GH suppresses ATM kinase activity, leading to decreased phosphorylation of key DDR proteins, including p53, Chk2, and H2AX. These effects were associated with decreased DNA repair by both NHEJ and HR as well as accumulation of unrepaired DNA damage. By contrast, blocking GH signaling with the GHR antagonist pegvisomant or with shGHR RNA interference in normal human colon cells resulted in increased ATM phosphorylation and decreased levels of endogenous DNA damage (84).

In vivo studies.

Most in vivo studies demonstrate DNA-damaging effects of GH, likely because of the benign nature of experimental models. Thus, in patients with acromegaly, measurement of ROS using plasma 8-OHdG showed a 2-fold increase in DNA damage compared with control subjects, and cytokinesis-block micronucleus assay showed increased peripheral lymphocyte chromosomal DNA damage (81, 90). Rats overexpressing a GH transgene as well as patients with acromegaly showed similarly increased ROS (91). Excessively high GH levels may not be necessary for this effect; administration of modestly supraphysiological doses of recombinant hGH to healthy volunteers for 3 weeks resulted in a higher percentage of heavily damaged lymphocyte nuclei as well as more lymphocyte chromosomal aberrations, with 3 of 9 individuals demonstrating persistently elevated DNA damage 30 days after hGH withdrawal (82). Furthermore, in a zebrafish model of acromegaly exhibiting a robust increase in the number of DNA-damaged cells assessed by γH2AX, gene set enrichment analysis showed impaired hepatic DNA repair pathways, including NHEJ and HR (83).

DNA-damaging effects of GH are also indirectly supported by studies of murine models of GH deficiency, which typically show ~30% increased lifespan (92). Protection from DNA damage was observed in GHR-knockout mice (93) as well as in GH-deficient long dwarf Snell mice deficient in GH due to spontaneous mutation in Pit1, responsible for somatotroph differentiation (94). As in the Pappa-knockout mice described above, these animals have diminished mTOR activity, which induces the DNA repair proteins MGMT and NDRG1 (70). Dwarf Ames mice harboring a mutation in PROP1 also responsible for somatotroph differentiation, serve as another model of longevity (95). GH treatment of Ames mice from 2 to 6 weeks of age countered their natural resistance to oxidative stress, resulting in decreased lifespan compared to WT mice, as well as significantly diminished resistance to DNA damaging chemicals in fibroblasts (96). Resistance to DNA alkylating agents and UV light were also observed in fibroblasts derived from young Snell mice (97), while prepubertal GH exposure of Lewis GH-deficient dwarf rats and Snell dwarf mice resulted in dysregulation of DNA damage proteins and decreased DNA repair efficiency (85). Similarly, our studies showed fewer SSBs and DSBs in colon tissue of GHR-knockout mice compared to WT littermates, while WT mice bearing GH-secreting xenografts exhibited increased levels of colon DNA damage after 5 weeks (84).

Results of studies in humans further demonstrate a role for GH in exacerbation of DNA damage. Non–GH-deficient short-stature children treated with GH for 6 to 12 months showed increased chromosomal breakage in lymphocytes upon treatment with bleomycin, and GH treatment of nontransformed lymphocytes amplified DNA-damaging effects of bleomycin (60). In addition, serum derived from GH-receptor-deficient human subjects reduced DNA breaks but increased apoptosis in human mammary epithelial cells treated with hydrogen peroxide, which creates ROS-induced DNA damage (98). These findings raise the question of whether DNA damage resistance contributes to the remarkable cancer resistance observed in patients with Laron syndrome harboring an inactivating GHR mutation (98) as well as longevity observed in murine models of GH signaling deficiency (92).

Unanswered questions.

Pro-proliferative and antiapoptotic effects of GH were recently reviewed (99), but data on the involvement of GH/GHR pathway in DDR are still limited, and several questions remain unanswered. For example, GH has been shown to protect intestinal mucosa from radiation (100-102), yet GH induces DNA damage in normal colon cells and tissue. GH suppresses p53, thus limiting apoptosis (103), and evidence suggests that acute antiapoptotic GH properties may underlie its protective effects, despite the presence of DNA damage. Over the long term, however, accumulated DNA damage ultimately results in cell transformation and an acquired tumorigenic phenotype, as evidenced by increased colony formation in normal human colon cells treated with GH and by mice bearing GH-secreting xenografts exhibiting increased numbers of metastases after a prolonged duration (84).

p53 status may account for the apparent discrepancy between GH effects in cancerous and nontransformed cells. Cancer cells undergo selection for improved survival and may rely on a compensatory DDR pathway if a single repair pathway (e.g., p53) is dysfunctional. This may, in turn, contribute to resistance to DNA-damaging chemotherapy and radiotherapy (104). Indeed, mutated p53 has been found in 43 of 56 colon cancer cell lines (105). Although GH induced DNA damage and suppressed DNA repair in nontumorous human colon and mammary cells (84), these effects were blunted in colon adenocarcinoma HCT116 cells (84) where p53 is mutated (106).

IGF1 may also contribute to opposing effects of GH in cancerous and nontransformed cells. GH suppression of p53 and induction of DNA damage in normal colon tissue occurs independent of IGF1 (84, 107). However, in cells and tissues where GH activates IGF1, IGF1 may counteract GH action on genome stability, inducing DNA repair and decreasing radio- or chemotherapy sensitivity. In acromegaly patients, and in murine models, GH excess is accompanied by increased IGF1, while GH deficiency is usually congruent with attenuated IGF1. A single publication reported that IGF1 deficiency in Pappa-knockout mice results in increased expression of DNA damage repair proteins (70). As IGF1R heterozygous mice demonstrate increased survival under oxidative stress (71), IGF1 may exacerbate DNA damage. Therefore, we cannot exclude that, in vivo, GH and IGF1 act concomitantly to induce DNA damage. If so, beneficial effects of GH deficiency on health and life span may also be mediated, at least in part, by decreased IGF1. However, evidence supporting this notion is quite limited.

Cell and tissue type may also determine direct effects of GH on DDR and DNA repair mechanisms. For example, GH suppresses p53 in colon, but not in liver (103), while p53 was upregulated in white adipose tissue in mice overexpressing the GH transgene (108). These results suggest that cell- and tissue-specificity are likely determinants of GH signaling on DDR and subsequent biological outcomes.

Proopiomelanocortin (POMC)/αMSH/ melanocortin 1 receptor (MC1R) signaling

Table 3 lists in vitro and in vivo studies related to the effect of αMSH on DNA damage and repair (Figure 4).

Table 3.

Effect of MSH/MC1R Signaling on DNA Damage and Repair

Author, Year (Reference)	Type of Study	Cells or Tissues	MSH/MC1R Manipulation	Type of DNA Damaging Treatment	Effects on DNA Damage or Repair	Mechanisms
Kadekaro et al., 2012 (116)	Ex vivo	Adult and neonatal human melanocytes	αMSH treatment	UV irradiation H2O2	↓ DNA damage	↑ ATR ↑ DNA-PKcs↑ Phospho-p53 ↑ GADD45 ↑ pChk2 ↑ OGG1 ↑ APE1 (BER pathway)
Swope et al., 2014 (119)	Ex vivo	Human melanocytes	αMSH treatment	UV irradiation	↑ DNA repair	↑ pATM ↑ pATR ↑ pChk1, pChk2 ↑ XPC (NER pathway)
Bohm et al., 2005 (117)	In vitro	Human melanocytes	αMSH treatment	UVB irradiation	↓ DNA damage ↑ NER
Dong et al., 2010 (121)	Ex vivo	Human primary keratinocytes	αMSH treatment	UVB irradiation	↓ DNA damage ↑ NER	XPA nuclear translocation
Smith et al., 2008 (123)	In vitro Ex vivo	B16 mouse melanoma cells, primary human melanocytes	MCR1R siRNA	UV irradiation	↑ DNA damage	↓ NR4A-dependent cyclobutene pyrimidine dimer removal (NER pathway)
Jarrett et al., 2017 (120)	Ex vivo	Primary human melanocytes	αMSH treatment	Cisplatin	↓ DNA damage	Enhanced XPA-pATR association and cAMP-dependent DNA repair
	In vitro	Human embryonic kidney (HEK293) cells	MC1R transfection		↓ DNA damage
Song et al., 2009 (118)	Ex vivo	Primary human melanocytes	αMSH treatment	UV irradiation	↓ DNA damage	↓ Oxidation MC1R activation

Abbreviations: MC1R, melanocortin 1 receptor; MSH, melanocyte stimulating hormone; NER, nucleotide excision repair; p, phosphorylated; siRNA, small-interfering RNA; XP, xeroderma pigmentosum.

Figure 4.

αMSH effect on DDR and DNA repair mechanisms. Shown in black are mechanisms known to be affected by αMSH (seeTable 3). αMSH increases phosphorylation and activation of key proteins in the DDR that block cell-cycle progression and slow propagation of damaged cells, and activate key DNA repair mechanisms that enable restored cell-cycle progression following DNA repair (seeFig. 1B). Abbreviations: ATM, ataxia-telangiectasia-mutated; ATR, ataxia telangiectasia and Rad3-related; BER, base excision repair; BRCA, breast cancer; DNA-PKcs, DNA-dependent protein kinase, catalytic subunit; HR, homologous recombination; NER, nucleotide excision repair; NHEJ, nonhomologous end joining; OGG1, 8-oxoguanin-DNA glycosilase; p53BP1, p53 binding protein-1; PARP, poly(ADP-ribose) polymerase; TERT, telomerase reverse transcriptase.

Involvement of the POMC/αMSH/MC1R signaling pathway in DDR emerged from studies of UV-induced DNA damage in the skin. POMC is synthesized in the skin, primarily in melanocytes and keratinocytes, and is processed into αMSH and adrenocorticotropin (ACTH) (109-111); exposure to UV light induces POMC, ACTH, αMSH, and MC1R expression in melanocytes and keratocytes (112-114), and a loss-of-function mutation in MC1R sensitizes melanocytes to DNA-damaging effects of UV radiation (115). MC1R, when activated either by αMSH or ACTH, triggers p53 phosphorylation and stabilization in human melanocytes, and promotes DNA repair by BER (116). αMSH, acting through MC1R, reduces UV-induced oxidative DNA damage (117, 118). Melanocyte treatment with αMSH induces xeroderma pigmentosum (XP) complementation group C (XPC), an enzyme that recognizes sites of DNA damage, enhances phosphorylation of ATR and ATM kinases and their respective substrates Chk1 and Chk2, and increases H2AX phosphorylation, thereby facilitating DNA repair (119). In melanocytes and human embryonic kidney HEK 293 cells transfected with MC1R, αMSH acting through cAMP enhances ATR phosphorylation and promotes ATR interaction with NER factor XP group A (XPA) to accelerate DNA repair and reduce mutagenesis of platinum-induced DNA damage (120); in keratinocytes, αMSH induces nuclear translocation of XPA to the DNA damage site (121). Nuclear receptor subgroup NR4A has also been implicated in αMSH regulation of DNA repair. NR4A2 is recruited to the DNA damage site in response to UV radiation to enhance NER (122), and MC1R signaling induces NR4A subgroup expression in human melanocyte and B16 mouse melanoma cells (123).

Although these studies were performed in vitro, in most studies, primary human melanocyte or ex vivo keratinocyte cultures were used, adding significance to the findings. The importance of POMC/MSH/MC1R signaling in DNA damage repair was further supported by in vivo results in humans showing requirement of fully functional MC1R for oxidative DNA damage control and protection from melanoma development (115, 124).

UV induction of POMC/αMSH in mouse and human skin may be controlled by p53 (125), as UV light generates DNA damage, activating p53. Acting as a transcription factor, p53 directly binds to the POMC promoter, inducing POMC transcription and αMSH production, which, in turn, activates DDR and DNA repair. In this way, POMC and αMSH appear to protect skin cells from DNA-damaging effects of UV light, preventing melanoma development, while mice devoid of p53 have increased propensity for developing UV-induced skin cancer (125, 126).

Gonadotropin-releasing hormone (GnRH)/Gonadotropin signaling

GnRH stimulates release of LH and FSH upon binding to the pituitary GnRH receptor (GnRHR), mediating gonadal functions. In murine LβT2 gonadotroph cells, GnRH induced mRNA and protein expression of FANCA, a DNA repair gene in the FA pathway. FANCA expression was also observed in adult mouse pituitary (127). GnRH enhances transient nuclear accumulation of FANCA protein in LβT2 cells, enabling GnRH-induced activation of glycoprotein hormone alpha-subunit (αGSU) and GnRHR gene promoters as well as transduction of GnRH signaling (128). These mechanisms may, at least in part, explain the reduced fertility rates seen in patients with inherited FA, characterized by mutated FANCA (129, 130).

The link between GnRH signaling and DDR also was evidenced by reports that administration of GnRH agonist during chemotherapy is beneficial for fertility preservation (131, 132). Further, experiments with murine ovarian fragments showed that LH, acting through cAMP, enhances repair of cisplatin-induced DNA damage assessed by the percentage of cells expressing H2AX (133). However, in rat Leydig cells, LH resulted in increased intracellular ROS, associated with increased DNA damage as assessed by Comet assay (134). As the number of studies in this area is limited, a full understanding of these mechanisms remains unclear.

Protective effects of FSH on sperm DNA fragmentation have also been described (135). In one study, infertile male patients were treated with FSH 3 times a week for 3 months, and sperm were analyzed using H2AX levels as an indicator of DNA damage before and after treatment. Functional improvement of sperm integrity was demonstrated, resulting in improved assisted reproductive outcomes with in vitro fertilization (136).

Reactive Hormonal Responses to DDR Pathway Mutations: Implications for Premature Aging

While a growing body of evidence indicates that peptide hormones affect DDR activation and damage repair, studies of DDR mutations in humans and animal models suggest that DDR may also manipulate endocrine functions and may manifest as endocrine disorders. Because these mutations are rare, cohort sizes small and studies mostly retrospective, the available information is observational and mostly limited to patients who participate in rare disease registries.

Mutations in 114 genes involved in DDR have been described (137, 138), and several rare diseases, including Seckel syndrome, FA, Nijmegen breakage syndrome, and Bloom syndrome, collectively classified as chromosomal instability syndromes (CIS), are characterized by increased chromosomal breakage resulting from unrepaired or misrepaired SSBs or DSBs. These progeroid syndromes (139) include patients exhibiting increased propensity for cancer and premature aging, with associated intrauterine growth retardation (140). However, each progeroid syndrome exhibits unique phenotypic biological features of aging, as DNA damage is stochastic and tissue-specific levels of DNA damage may vary due to heterogeneous DNA repair mechanisms (141).

To date, endocrine alterations have only been reasonably evaluated in patients with FA. DNA ICLs that affect transcription and DNA replication are corrected by the FA/BRCA DNA repair pathway. In FA, malfunctioning of any of more than 20 FA proteins results in accumulated DNA damage and the clinical FA phenotype of short stature, congenital malformation of upper extremities, head, ears, and kidney, bone marrow failure, and increased cancer susceptibility (140, 142, 143). Importantly, although clinical features vary depending on the specific FA gene involved, 80% of patients exhibit endocrine alterations (140), including low birth weight, GH deficiency, hypothyroidism, and hypogonadism (144). About half of these patients show short stature, often attributed to GH deficiency, but which may also be a consequence of hypothyroidism due to low TSH levels (142-144). Small pituitary size was reported in 10 of 11 patients (145), but this finding was not confirmed (144). Histological findings, albeit limited, suggest that hypogonadism and other reproductive abnormalities may result from primary gonadal defects, with hypothalamic-pituitary dysfunction responsible for reduced fertility (130). Reduced fertility was also observed in a FA animal model with ovarian hypoplasia and disrupted spermatogenesis (146), but endocrine evaluation was not reported.

DNA damage may underlie the endocrine phenotype of FA and other CIS patients in several ways. Intact FA proteins are required for transcriptional effects of GnRH on pituitary gonadotroph hormone genes (128), and these mutations may be associated with reduced fertility observed in FA (129, 130). The limited evidence would suggest that as unrepaired DNA damage accumulates, increased apoptosis may eliminate pituitary or peripheral hormone-producing cells. Alternatively, endocrine cells could also be damaged by high levels of ROS, known to be elevated in FA (142). Finally, a decline in pituitary GH-secreting somatotroph numbers may result from exquisite sensitivity of GH-producing cells to DNA damage compared with other hormone-producing pituitary cell types (147).

Accumulation of DNA damage and chromosomal instability, hallmarks of aging (148), explain the progeroid features of FA and other CIS. Premature aging associated with DDR is also seen in Cockayne syndrome (CS, affected genes CSA or CSB) and XP (affected genes XPA, XPB, XPD), both of which show defective NER DNA repair mechanisms. In patients with CS, transcription-coupled NER (TCR), which removes DNA lesions that obstruct RNA polymerase, is defective, while global genome (GG) NER, which repairs DNA damage in both transcribed and untranscribed DNA strands, is mutated in patients with XP (149). Patients with XP-CS complex have defects in both TCR and GG-NER.

CS patients display postnatal growth failure, progressive neurological abnormalities, hearing loss, and impaired sexual development. Very few affected individuals have undergone thorough endocrinological examination, and results of these studies are mixed, with circulating GH levels reported as low in some patients and normal in others (150, 151). In mutant Csa^m/^mXpa^-/^- mice serving as an animal model of CS, inactivation of NER caused a phenotype reliably mimicking CS syndrome, with premature aging and early death. Whole genome transcriptome analysis of the animals revealed suppression of the GH/IGF1 somatotroph axis. These changes appeared to not originate from pituitary dysfunction but were manifested by decreased expression of liver Ghr, Igf1, and Igfbp3, as well as energy metabolism genes (152). Similarly, progeroid Ercc1^-/^- mice, a model of XP, showed downregulated GH/IGF1 signaling, including decreased liver mRNA levels of Ghr, Igf1, Igfbp, and Prlr, despite increased levels of circulating GH indicative of unaltered pituitary function and perhaps GH resistance. Experiments in NER-mutant primary mouse fibroblasts and human NER-defective fibroblasts showed that arrest of RNA polymerase II in persistent DNA lesions resulted in attenuated IGF1R and GHR, while experimental removal of DNA lesions alleviated these changes (153). These findings suggest that accumulation of unrepaired DNA damage likely underlies observed suppression of GH signaling in these models (152).

We interpret these results from a mechanistic point of view. DNA damage upregulates p53, as observed in animals with DNA repair deficiencies (e.g., Ercc-1^-/- mice) showing elevated levels of p53 (154). In turn, induced p53 transcriptionally suppresses IGF1R (155), as reported in fibroblasts derived from DDR mutant mice (153, 156). Concurrently, DNA-damage-induced p53 stimulates the GH promoter in pituitary and in nonpituitary cells (34), thus enabling GH upregulation not only in the pituitary, but also in local peripheral tissue where locally induced GH may exacerbate DNA damage (84). Although local tissue GH was not measured in DDR mutant animals, elevated levels of circulating GH were reported (152, 156). Declines in liver Ghr and Igf1 abundance and resultant growth arrest may therefore reflect an attempt to protect tissues against DNA damaging effects of induced GH. Overall, the evidence suggests that extensive DNA damage in DDR mutant mice, not only in the liver but also in other tissues, appears to override this protection, resulting in premature aging.

Changes in somatotroph axis observed in prematurely aging DDR mutant models strikingly recapitulate changes observed in aging humans that exhibit somatopause (157-159), as well as in aging animals (160, 161). In progeroid disorders, aging features are evident in multiple organs (162). Global gene expression in primary murine cells undergoing persistent DNA damage was similar to that occurring in naturally aged animals (153). Fibroblasts derived from patients with Hutchinson-Gilford progeria syndrome exhibit changes in activity of multiple signaling pathways, with induced ERK, mTOR, and MAPK, downregulation of DNA repair, altered chromatin organization, and mitochondria dysfunction similar to that seen in cells derived from aging individuals (163, 164). In senescent cells, DNA damage accumulates due to defective DNA repair (30), and elimination of senescent cells in the BubR1 progeroid mouse delays age-related pathologies in the eye, adipose tissue, and skeletal muscle (165). Thus, although circumstantial, evidence suggests that inadequate DNA repair and DNA damage accumulation may be causally linked to premature aging.

It has been suggested that suppressed GH/IGF1 somatotroph axis in DNA-damage repair deficient models as well as in aging organisms is a component of a conserved pathway that shifts energy usage away from growth and proliferation and toward activities that minimize DNA damage, maintain organismal integrity, and potentially prolong life span (153, 156, 166). In addition, attenuation of somatotroph axis was proposed to also prevent damaged cells from transformation into tumor cells by suppressing growth-stimulating signals (162). During the aging process, DNA damage accumulates in senescent cells located in normal tissues, activating local GH, which, by suppressing DDR and damage repair, further exacerbates DNA damage. Accordingly, in DDR mutant mice, attenuated GH/IGF1 signaling decreased levels of GHR and IGF1, conferring protection from the effects of local GH.

At this time, available evidence suggesting protective effects of somatopause in aging are largely indirect. However, GH/IGF1 signaling deficiency in animal models is associated with a longer and healthier life span (167-172). Mice with abrogated GHR function live longer and are intrinsically resistant to cancer (99, 173) and patients with the inherited GHR deficiency Laron syndrome do not develop cancer (98). Individuals with partial IGF1 resistance exhibit exceptional longevity (72, 73). Studies of GHR-knockout mice revealed decreased rates of endogenous DNA damage, implying that, in the absence of GH, more effective DNA repair mechanisms may underlie cancer resistance and longevity (84). Somatopause in humans may therefore be viewed as protecting against acquisition of tumorigenic mutations and other diseases associated with DNA damage accumulation and aging (157, 174).

Future Directions

Defects in DNA maintenance and repair resulting in destabilization of copy number and nucleotide sequence has emerged as an enabling characteristic for tumor development. Recent reports highlight the importance of peptide hormone signaling in maintaining DNA and chromosomal stability. Although evidence is still incomplete and observations sometimes inconsistent, it is apparent that hormonal status such as in acromegaly or with GH, IGF1, or POMC deficiency likely has a significant effect on DNA and genomic integrity. Conversely, suppressed DDR or accumulated DNA damage can have a profound impact on hormonal status as evident in patients harboring mutations in DNA damage response or repair pathways, as well as in animal models of these diseases. These observations have enhanced our knowledge of peptide hormone physiologic actions and add new perspectives to our understanding of mechanisms protecting genome integrity or driving the aging process.

Several questions remain unresolved. The remarkably complex DNA damage repair process encompasses different signaling pathways with transcription, translation, and posttranslational modifications. Precise mapping of mechanisms underlying sites of hormonal actions are necessary to fully understand the impact, but it is yet uncertain whether all observed effects in cell-based systems will prove relevant in vivo.

Furthermore, some proteins function differently in nontransformed and in neoplastic cells, while, for others, there is insufficient evidence to fully evaluate their respective contribution. Thus, although a few well-established studies show that POMC-derived αMSH acting through MC1R enhances DNA repair and provides a barrier for potential harmful skin exposure to UV light, to our knowledge, there is limited information on how POMC/αMSH signals in melanoma cells.

Differences in polypeptide hormone action in nontumorous and tumorous cells may also depend on p53 status. For example, exposure of normal keratinocytes to UV light results in p53 activation with subsequent POMC/αMSH induction and activated DNA repair (125). In many cancer cells, p53 is mutated, and loss of p53 function may therefore underlie a well-described association between p53 mutation and skin cancer (175). The ultimate effects of hormones on DNA repair may also depend on intermediate interacting factors modulating their action. Thus, PTEN was shown to play an essential role in NER DNA repair by promoting XPC transcription (176). In nontumorous human colon epithelial cells, GH attenuates PTEN expression (103), likely exacerbating suppressive effects of GH on DNA repair.

Age may also play a role in hormone action on DNA damage. IGF1 and IGF1R are required for activation of Chk1 kinase by ATR after UV irradiation. IGF1R activity is significantly attenuated in geriatric human keratinocytes, resulting in abrogation of NER and potentially leading to age-associated skin carcinogenesis (59).

The IGF1/IGFR pathway appears to universally enable DNA damage repair, yet GH effects on DDR differ significantly by cell type. In tumor cells, both GH and IGF1 enhance DNA damage repair, as observed by GH protecting malignant cells against DNA damaging chemotherapy, prompting the suggestion of using GH inhibitors as an adjuvant treatment. However, we have to be mindful of other effects of GH, including effects on insulin resistance and cardiovascular targets. Furthermore, in nontransformed cells and in normal murine colon tissue, GH enhances both endogenous and etoposide-induced DNA damage, suppressing DNA repair, and these effects occur independent of IGF1. Adding another layer of complexity, GH induced in senescent tumor cells in response to DNA damaging therapy could, through a paracrine effect, suppress DDR in neighboring cells and reprogram transformation of normal tissue.

GH-related products have been aggressively promoted as antiaging therapy (174, 177) and are widely used in an attempt to enhance athletic performance (178, 179). As GH enhances accumulation of unrepaired DNA damage in normal tissues, there is a concern that long-term GH treatment may lead to a risk of developing epithelial cell transformation, as well as potential activation of underlying low-grade tumors. That said, physiologic GH replacement is efficacious and generally safe when administered to children and adults with GH deficiency and short stature or adult GH deficiency syndrome (174). GH recipients should be monitored as adults and may require continued low-dose GH administration if consequent adult GH deficiency is confirmed (180).

Understanding these dynamic variations should elucidate the role of peptide hormones in the intricate machinery of DNA damage response and repair and will expand our knowledge of how to harness these pathways for therapeutic advantage.

Glossary

Abbreviations

8-OHdG

8-hydroxy-2’-deoxyguanosine

αMSH

α-melanocyte stimulating hormone

ACTH

adrenocorticotropin

ATM

ataxia-telangiectasia-mutated

ATR

ataxia telangiectasia and Rad3-related

BER

base excision repair

CIS

chromosomal instability syndromes

CS

Cockayne syndrome

DDR

DNA damage response

DNA-PKcs

DNA-dependent protein kinase, catalytic subunit

DSB

double strand break

FA

Fanconi anemia

GH

growth hormone

GHRH

growth hormone–releasing hormone

GnRH

gonadotropin-releasing hormone

H2AX

histone H2 variant X

HR

homologous recombination

ICL

interstrand crosslink

IGF1

insulin-like growth factor 1

IGF1R

insulin-like growth factor 1 receptor

IGFBP

IGF binding protein

MC1R

melanocortin 1 receptor

miRNA

microRNA

mTOR

mammalian target of rapamycin

NER

nucleotide excision repair

NHEJ

nonhomologous end joining

NR4A

nuclear family 4 subgroup A receptor

OGG1

8-oxoguanin-DNA glycosilase

PARP

poly(ADP-ribose) polymerase

POMC

proopiomelanocortin

ROS

reactive oxygen species

siRNA

small interfering RNA

SSB

single strand break

WT

wild-type

XP

xeroderma pigmentosum

Additional information

Disclosure Summary: The authors have nothing to disclose.

Data Availability: Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.