TABLE 1.
Summary of in vitro, animal and human studies pointing towards an involvement of accelerated aging in CF disease.
| In vitro studies | |||
|---|---|---|---|
| Hallmark of aging | Cell culture model | Main findings | Ref |
| Genomic instability | Human fetal lung fibroblast cell line IMR-90, primary human bronchial epithelial cells, and lymphocytes isolated from human blood | ROS and proteases from activated neutrophils increase ROS in mitochondria and cytoplasm, which is associated with oxidative cell injury and death | (Aoshiba et al., 2001) |
| Human CF lung epithelial cell line IB3-1 (compound heterozygot for ∆F508 and 128W2X) and isogenic stably wild-type (wt) CFTR transfected C38 cells | Decreased mitochondrial reduced glutathione (GSH) and increased ROS in CFTR deficient human lung epithelial cells | (Velsor et al., 2006) | |
| CFTR overexpression and knockdown in A549 cell line | Inhibition of CFTR activity promotes epithelial-mesenchymal transition through the uPA/uPAR pathway | (Li et al., 2015) | |
| Loss of proteostasis | COS-7 cell line | Mutated CFTR such as ∆F508 remains in the endoplasmatic reticulum (ER) and is sequester for degradation | (Cheng et al., 1990) |
| Calu-3 (wt and stably ∆F508-CFTR transfected) and CFPAC-1 (endogenous ∆F508-CFTR) cell line | ∆F508-CFTR overexpression causes ER stress and activates the unfolded protein response leading to decreased wt CFTR mRNA and protein maturation | (Bartoszewski et al., 2008) | |
| Normal and CFTR mutated CFBE41° bronchial epithelial cells, primary human bronchial/tracheal epithelial cells and HeLa cells | Functional CFTR controls its own surface expression in a positive feed-forward loop through its effects on the proteostasis network. siRNA depleting CFTR interferes with endosomal trafficking of cell surface proteins. Proteostasis regulator cystamine corrects the deranged proteostasis | (Villella et al., 2013) | |
| FRT cell line, HEK-293 cells, and primary human bronchial epithelial (HBE) cells | CFTR corrector VX-809 improves F508del-CFTR processing in the ER, leading to plasma residence time and susceptibility to proteolysis similar to normal CFTR. | (Van Goor et al., 2011) | |
| Human normal and CF bronchial epithelial cell lines (CFBE41o-, IB3-1, 16HBE14o-), ex-vivo cultures of nasal polyp mucosal biopsies and brushed nasal epithelial cells from ∆F508 homozygous patients and matched controls | Proteostasis regulators cystamine and EUK-134 (superoxide dismutase/catalase-mimetic) improve ∆F508-CFTR trafficking and stability at the epithelial cell surface by overexpressing BECN1 and depleting SQSTM1. This facilitates its response to CFTR potentiators and suppresses inflammation | (Luciani et al., 2012) | |
| IB3-1 and isogenic stably rescued C38 cells and peripheral blood mononuclear cells (PBMCs) from pediatric CF patients and healthy controls | Dysfunctional autophagy appears to contribute to the exaggerated CF lung inflammation. Improving autophagosome clearance attenuates the hyperinflammatory response | (Mayer et al., 2013) | |
| IB3-1 and isogenic stably rescued C38, 16HBE and A549 cell line, ex vivo cultures of nasal polyp mucosal biopsies from CF patients and controls | Defective CFTR function generates oxidative stress that leads to PIASy mediated tissue trans-glutaminase 2 (TG2) SUMOylation inhibiting its ubiquitination and proteasome degradation. TG2 inhibition increases NF-κB inhibitor Ikκα | (Luciani et al., 2009) | |
| Deregulated nutrient sensing | CHO (wt and stably expressing CFTR), T84 and Calu-3 cell line, and Xenopus oocytes | AMPK and CFTR are endogenously expressed in the same tissue types and have been found to interact with each other leading to CFTR phosphorylation and altered CFTR Cl− conductance. This may represent a link between transepithelial transport and cell metabolic state | (Hallows et al., 2000) |
| CF human bronchial epithelial cell line CFBE41o- (∆F508 mutation) and isogenic HBE41o- cells (wt CFTR) | ∆F508-CFTR interactome differs highly from its wt counterpart including differences in the mTOR, JAK/STAT and several other pathways, showing the catastrophic effects from one misfolded protein on protein-protein interactions | (Pankow et al., 2015) | |
| CFBE41o- and HBE41o- cells | Inhibition of the PI3K/Akt/mTOR pathway leads to improved CFTR stability, while select inhibitors of this pathway leads to restored autophagy and reduced ∆F508-CFTR aggregates | Reilly et al. (2017) | |
| CFBE41o- and 16HBE14o- cells | Reduced level of transcription factor FOXO1 and β2 arrestin, along with increased ERK1/2 in CF cells. FOXO1 reduction is linked to loss of CFTR function and increased after insulin-like growth factor 1 (IGF-1) administration. Reduced FOXO1 may explain insulin insensitivity in CF, with IGF-1 constituting a potential treatment of CF-related diabetes | (Smerieri et al., 2014) | |
| CFBE41o-, 16HBE14o-, and IB3-1 cells | Altered transcriptional profile of miRNAs in CF cells, four of which are potential FOXO1 regulators. These four miRNAs are also differentially expressed in CF patients, and dependent on genotype and glucose tolerance state. This may explain some of the variability in metabolism among CF patients | (Montanini et al., 2016) | |
| Mitochondrial dysfunction | IB3-1 cell line | Pseudomonas eruginosa induced inflammation shows importance of mitochondria in the pro-inflammatory condition in CF including their role in Ca2+ signaling along with NLRP3 recruitment and activation | Rimessi et al. (2015) |
| CF and non-CF HBE cells | ∆F508-CFTR correctors recover diminished function of the major redox balance and inflammatory signaling regulator Nrf2, inducing its nuclear translocation and transcription of target genes. Nrf2 rescue is dependent on CFTR function | (Borcherding et al., 2019) | |
| Cellular senescence | Normal HBE cells | Neutrophil elastase (NE) triggers cell cycle arrest through elevated p27Kip1 expression resulting in G1 arrest in normal HBE cells | (Fischer et al., 2007) |
| Normal HBE cells | NE increases p16 expression and decreases CDK4 activity in HBE cells, which may explain how NE treatment triggers cell cycle arrest | (Fischer et al., 2013) | |
| Stem cell exhaustion | HBE cells | No general telomere shortening in CF HBE cells leading to the conclusion that progenitor reserve is sufficient to maintain normal telomere length despite enhanced cell turnover | (Fischer et al., 2013) |
| IB3-1 and control CFTR repaired IB3-S9 cells | CF lung epithelial cells hyperexpress miRNA-155, also upregulated in aging. This activates PI3K/Akt signaling through reduced SHIP1. Resulting activation of downstream MAPKs stabilizes IL-8 mRNA and thus increases IL-8 expression promoting inflammation | (Bhattacharyya et al., 2011) | |
| Altered cellular communication | NCI-H441 and 16HBE14o- cells | Functional CFTR downregulates NF-κB activity. CF associated hyper-inflammation may represent a consequence of insufficient inhibition of NF-κB signaling | Hunter et al. (2010) |
| HBE cells | TGF-β1 decreases expression of the γ-subunit LRRC26 of the apically located large-conductance Ca2+- and voltage-dependent K+ (BK) channels. Thereby, TGF-β1 reduces BK activity, airway surface liquid volume and ciliary beat frequency | Manzanares et al. (2015) | |
| Normal and homozygous △508-CFTR HBE cells | TGF-β1, which is frequently elevated in CF patients, reduces CFTR mRNA and protein level in non-CF HBE cells. TGF-β1 also impairs functional rescue of △508-CFTR suggesting it may interfere with therapies aiming at correcting the processing defect of △508-CFTR. | Snodgrass et al. (2013) | |
| T84 cell line and HBE cells | TGF-β reduces calcium activated chloride conductance (CaCC) and CFTR-dependent chloride currents. It reduces expression and activity of TMEM16A and CFTR, and reverses △508-CFTR correction by VX-809. Inhibition of Smad3 and p38 MAPK partially reverses TMEM16A and CFTR downregulation | Sun et al. (2014) | |
| NCI-H292 cell line infected with wt or △508-CFTR | NE promotes degradation of wt and △508-CFTR through activation of intracellular calpain protease causing loss of channel function | Le Gars et al. (2013) | |
| In vivo animal studies | |||
| Hallmark of aging | Animal model | Main findings | Ref |
| Genomic instability | Congenic CFTR KO strains (S489X and FABP) and C57B6 control mice | Decreased mitochondrial GSH in lungs of CFTR-knockout mice, and increased mitochondrial DNA oxidation and oxidative stress | Velsor et al. (2006) |
| Athymic balb/c mice injected with CFTR knockdown or control A-549 cells | CFTR status affects cell invasion and migration. No difference in primary tumor growth between control and CFTR-knockdown A549-injected mice, but increased lung metastasis and increased tumor burden by CFTR-knockdown cells | Li et al. (2015) | |
| Telomere attrition | Nfkb1-/- mice (C57B1/6 background) and fibroblast cultures from ear clippings, p55△ns/△ns mice, and late-generation (F3-F4) terc−/- mice bred from B6/Cg-TERCtm1Rdp/J | Chronic, low-grade inflammation induces telomere dysfunction, senescence, impaired tissue regeneration and premature aging. Conversely, telomere dysfunction leads to a pro-inflammatory state. Anti-inflammatory or antioxidant treatment blocks accumulation of telomere-dysfunctional senescent cells in nfkb1-/- tissues | Jurk et al. (2014) |
| Loss of proteostasis | CftrF508del/F508del mice (129/FVB outbred background) and wt littermates | Autophagy restoration by cystamine treatment, BECN1 overexpression and SQSTM1 depletion all considerably increase ∆F508-CFTR at the epithelial surface and decrease lung inflammation | Luciani et al. (2012) |
| CftrF508del/F508del, cftr -/- and CftrF508/- mice, CftrF508/+ and CftrF508del/F508del mice in Becn1 haploinsufficient background (Becn1+/−) | Cystamine plus epigallocatechin gallate restore CFTR function and reduce lung inflammation in CftrF508del/F508del and CftrF508/- mice. ∆F508-CFTR rescue is linked to autophagy restoration, i.e., no rescue in Becn1+/− background | Tosco et al. (2016) | |
| CftrF508del/F508del mice (129/FVB outbred background) and wt littermates | TG2 inhibition by cystamine restores PPARγ levels and nuclear localization, and reduces TNFα in lungs of CF mice | Luciani et al. (2009) | |
| Deregulated nutrient sensing | Young adult CF mice homozygous for F508del-CFTR and wt litter mates | Reduced FOXO1 in muscle, but not in liver and adipose tissue of CF mice. Insulin-like growth factor 1 (IGF-1) increases FOXO1 in CF muscle tissue similar to the wt level, and increases it in adipose tissue of both mouse models | Smerieri et al. (2014) |
| Mitochondrial dysfunction | CF mouse models (Cftrtm1kth, Cftrtm2Mrc, and Cftrtm1Unc in C57BL/6J background) | Reduced Nrf2-CFTR colocalization in CF mouse models with concomitantly reduced expression of Nrf2 target geneses HMOX1, NQO1, and GCLC. | Borcherding et al. (2019) |
| Cellular senescence | Nfkb1-/- mice (C57B1/6 background) and fibroblast cultures from ear clipping, p55△ns/△ns mice, and late-generation (F3-F4) terc−/- mice bred from B6/Cg-TERCtm1Rdp/J | Chronic, low-grade inflammation induces telomere dysfunction, senescence, impaired tissue regeneration and premature aging. Anti-inflammatory or antioxidant treatment blocks accumulation of telomere-dysfunctional senescent cells in nfkb1-/- tissues | Jurk et al. (2014) |
| Altered cellular communication | Nfkb1-/- mice (C57B1/6 background) and fibroblast cultures from ear clippings, p55△ns/△ns mice, and late-generation (F3-F4) terc−/- mice bred from B6/Cg-TERCtm1Rdp/J | Chronic, progressive low-grade inflammation promoted by NF-κB can cause premature aging | Jurk et al. (2014) |
| C57/B16 and NE−/- mice | NE promotes CFTR degradation through activation of intracellular calpain protease causing loss of channel function | Le Gars et al. (2013) | |
| Human studies | |||
| Hallmark of aging | Study design | Main findings | Ref |
| Genomic instability | 20 years follow up of CF patients receiving care in United States CF care center programs comparing their cancer incidence with that of the general population | Increased risk for lymphoid leukemia, testicular cancer and digestive tract (esophago-gastric junction, biliary tract, small bowel and colon) cancers in non-transplanted patients compared to the general population. Particularly high risk for digestive tract (mostly bowel) cancers in transplanted patients | Maisonneuve et al. (2013) |
| Cancer incidence in CF and non-CF lung transplant recipients compared to the general population based on the United States transplant and 16 cancer registries | Overall cancer risk in CF lung transplant recipients is more increased than in non-CF recipients, particularly for colorectal cancer, esophageal cancer and non-Hodgkin lymphoma | Fink et al. (2017) | |
| CFTR expression in non-small cell lung cancer (NSCLC) and normal lung samples | Downregulated CFTR expression in NSCLC samples; correlation of low CFTR expression with advanced stage, lymph node metastasis and poor prognosis | Li et al. (2015) | |
| Urinary 8-hydroxydeoxyguano-sine (oh8dG, marker of free radical-induced DNA damage) from CF patients and age matched healthy controls and correlation with clinical status | Significantly higher urinary oh8dG in CF group, but no correlation with markers of lung function (FEV1, FVC) or clinical status assessed by Taussig-Schwachman score in CF patients; highly significant, positive correlation between urinary oh8dG and plasma vitamin E | Brown et al. (1995) | |
| Bronchoalveloar lavage (BAL) of CF and nonsmoking control subjects | Glutathione deficiency with significant reduction in GSH in CF BAL fluid. Simultaneously marked deficiency of plasma GSH. | Roum et al. (1985) | |
| BAL of children with CF and normal control subjects | Highest oxidative stress level assessed by protein carbonyls in patients with FEV1 < 80% of predicted or highly elevated neutrophils | Starosta et al. (2006) | |
| Telomere attrition | Telomere length of DNA extracted from airway epithelial cells of CF patients and controls | No significant difference in telomere length between CF and control airway epithelial cells, apart for a small subgroup of CF subjects showing shorter telomeres | Fischer et al. (2013) |
| Telomere length of DNA extracted from peripheral blood leukocyte of CF patients | Decreased telomere length associated with severe disease characterized by lower FEV1, ∆F508 homozygosity and CF asthma | Lammertyn et al. (2017) | |
| Telomere length of DNA extracted from tissue cores of healthy and diseased human lungs from (re-)transplantation or autopsy | Decreasing telomere length with age in normal, but not in CF lungs. No reduction in telomere length in CF explant lungs compared to normal lungs and no correlation between telomere length and disease severity | Everaerts et al. (2018) | |
| Epigenetic alterations | DNA methylation profiling in nasal epithelial cells and whole blood from CF and control subjects | Substantial DNA methylation differences between CF and control samples, and different methylation levels between mild and severe CF disease. DNA methylation changes in genes responsible for epithelial integrity, inflammatory and immune response, and over-represented in enhancers active in lung tissue | Magalhães et al. (2018) |
| DNA methylation profiling in BAL cells, i.e., primarily lung macrophages, from CF and normal subjects | Multiple differentially methylated CpG sites in CF BAL cells, mostly hypo-methylated and located in non-promoter CpG island as well as putative enhancer regions and DNase hypersensitive regions. Altered DNA methylation significantly associated with CF disease status and may be a driving factor in CF innate immune dysfunction | Chen et al. (2018) | |
| Loss of proteostasis | Single-center, open-label phase-2 clinical trial with CF patients orally treated with cysteamine and epigallocatechin | Beneficial effects of proteostasis regulators in patients with rescuable CFTR with decreased sweat chloride, increased CFTR expression and function, restored autophagy, decreased sputum CXCL8 and TNF-α and tendency for improved respiratory function | Tosco et al. (2016) |
| Cellular senescence | Immunohistochemistry for senescence markers on tissue sections of lungs from CF patients and normal controls | Significantly increased expression of senescence/DNA damage markers (p16INK4a, γH2A.X and phospho-Chk2) in CF airways | Fischer et al. (2013) |
| Stem cell exhaustion | Bronchial epithelial cells from lung brush biopsies and blood neutrophils from CF patients homozygous for ∆F508 mutation and normal controls | miRNA-155, upregulated in aging and suspected to contribute to inflammation-associated stem cell dysfunction (Teramura and Onodera, 2018), is highly expressed in CF lung epithelial cells and blood neutrophils, where it reduces SHIP1 levels promoting PI3K/Akt activation that drives IL-8 expression | Bhattacharyya et al. (2011) |
| Altered cellular communication | TGF-β1 gene polymorphism in CF patients homozygous for ∆F508 | Patients with high levels of the profibrotic, pro-aging cytokine TGF-β1 show accelerated lung function decline | Arkwright et al. (2000) |
| TGF-β1 level in BAL fluid of pediatric CF patients and non-CF controls, and assessment of clinical disease indicators | Elevated TGF-β1 in CF patients is associated with neutrophilic inflammation, diminished lung function and recent hospitalization. No association of TGF-β1 with presence or quantity of bacterial pathogens | Harris et al. (2009) | |
| Estimation of lung volumes and forced expiratory flows in children before clinically indicated BAL | TGF-β1 and matrix metalloprotease (MMP)-2, both previously linked to airway remodeling, are associated with diminished flows as well as hyperinflation and air trapping | Peterson-Carmichael et al. (2009) | |
| ECM dysregulation | Matrix metalloprotease (MMP) profile in lower airway secretions of CF in- and outpatients and normal controls | Increased active MMP-9 and NE, and decreased tissue inhibitor of metalloprotease-1 (TIMP-1) in CF. NE can activate pro-MMP-9 and degrade TIMP-1, hence not surprisingly a strong correlations between NE and MMP-9 activity in CF inpatient samples | Gaggar et al. (2007) |
| BAL and endobronchial biopsies in children with CF and controls without lower airway symptoms | Increased elastin, glycosaminoglycan and collagen in BAL fluid from children with CF indicating matrix breakdown that positively correlates with age, neutrophil count and protease concentration. Increased reticular basement membrane thickness in CF group correlates with TGF-β1 level in BAL fluid | Hilliard et al. (2007) | |