| Telomeres cap the chromosome ends and prevent the activation of DNA damage checkpoints inducing cell cycle arrest (senescence) or apoptosis. There is ample evidence that telomere shortening occurs during human aging and chronic diseases. In addition, mutations in the enzyme telomerase lead to impaired tissue maintenance and shortened lifespan in humans and mice. These genetic disorder show that telomere shortening can impair organ maintenance and shorten lifespan. However, the actual contribution of dysfunctional telomeres to natural human aging and diseases remains under debate. Accumulation of senescent cells has been detected in skin of aging humans and primates but not in others organs such as muscle or liver. In aging telomerase knockout mice (mTerc-/-) telomere dysfunction is associated with a decline in stem cell function, impaired organ maintenance, and a shortened lifespan. Yet, mTerc-/- mice do not show an accumulation of senescent cells. There is emerging evidence that senescent cells can be cleared in vivo by induction of apoptosis and immune responses. Moreover, the detection of senescence is cell cycle dependent17. Therefore, the impact of telomere dysfunction on aging might be underestimated by experiments trying to detect senescent cells. In addition, the detection of senescent cells in vivo is technically challenging impeding its implementation as a clinical marker. In vitro studies on human cells indicate that low levels of telomere dysfunction precede cellular senescence in culture. The identification of marker proteins indicating low levels of telomere dysfunction in pre-senescent cells could help to determine the influence of telomere dysfunction on human aging and disease. In addition, the identification of such markers is of clinical interest since there is growing population of humans suffering from disease phenotypes associated with aging.Since late generation of telomerase knockout mice (G4 mTerc-/-) do not show an accumulation of senescent cells but yet develop aging phenotypes associated withtelomere dysfunction, we used this mouse model to identify secreted marker proteins of telomere dysfunctional cells. We have analyzed peptides in supernatant of short term cultures (4 h) of total bone marrow cells from 2 vs. 12 months old mTerc+/+ and G4mTerc-/- mice using a proteomics approach (CE-TOF-MS). Statistical analysis followed by machine learning procedures identified and validated a set of peptides that distinguished telomere dysfunctional (G4mTerc-/-) mice from wild type (mTerc+/+) mice as well as 12 months old from 2 months old mice with high accuracy in a blinded test set consisting of 26 additional samples (Suppl. Fig.1a-e). In addition, we identified a set of peptides by tandem mass spectrometry sequencing analysis. Among them, four of these peptides showed increased expression in 12 months old G4mTerc-/- mice compared to all other 3 mouse cohorts (n=5 mice per group, figure 1a). The up-regulation of these markers in aged G4mTerc-/- mice was confirmed in a blinded test set of 26 mice: 9 out of 15 of G4mTerc-/- mice showed an up-regulation compared to 1 out of 11 mice of the 3 other cohorts (specificity: 91%, sensitivity: 60%). Western blot analysis on mouse bone marrow cells reconfirmed the up-regulation of these marker proteins in aging G4mTerc-/-mice (figure 1b).The sequenced peptides were fragments derived from following proteins: (i) Cathelicidine related antimicrobial protein (CRAMP) - it is activated during innate immune responses and protects against bacterial infection but has not been associated with aging, (ii) Chitinase 3 like protein 3 (Chi3L3) - it belongs to the chitinase gene family, which is activated in innate immune responses a member of this family hasbeen associated with chondrocyte aging and arthritis , (iii) Elongation factor 1α, (EF-1α) - it controls translational protein synthesis and is up-regulated during senescence of human fibroblasts, (iv) Stathmin (OP 18) - it leads to microtubule destabilization and has been implicated in the control of mitosis and cell motility.While mRNA expression of the identified markers was similar in 2 monthsold G4mTerc-/- and mTerc+/+ mice (figure 1c), an up-regulation of the markers was detected in various organs of 12 months old G4mTerc-/- mice but not in age-matched or 24 months old mTerc+/+ mice (Figure 1c). Some of the proteins (CRAMP, Ch3L3) were up-regulated in all analyzed organs (kidney, liver, lung, brain, spleen, heart), whereas others (EF-1α, Stathmin) showed organ specificity. Immunohistochernistry (IHC) confirmed the increased expression of the marker proteins in tissues of aging G4mTerc-/- mice but not in mTerc+/+ mice (figure 1d, Suppl. Table 1, Suppl. Fig.2). In addition, an up-regulation of all markers was observed in blood serum of 12 months old G4mTerc-/- mice but not in age-matched mTerc+/+ mice (figure le). Only CRAMP-protein level and chitinase enzyme activity showed a moderate up-regulation in 24 months old mTerc+/+ mice, but not as high as in 12 months old G4mTerc-/- mice. The expression of the protein markers was not affected in long-living mouse mutants (growth hormone receptor knockout mice, Ames dwarf mice) with alterations in growth hormone signaling (Suppl. Figure 3). Together, these results indicated that we had identified a set of marker proteins that is specifically associated with aging of telomere dysfunctional mice but not with other causes of aging in mice with long telomere reserves.Next, we examined the expression of the identified marker proteins in human aging. Antibodies were available for 3 orthologs (CRAMP, EF-1α, Stathmin). A human ortholog for mouse Ch3L3 has not yet been identified, but it is possible to measure chitinase enzyme activity across different species, and it is also possible to measure chitinase RNA level. In human blood serum, the expression levels of CRAMP (p<0.0001), EF-1α(p =0.0004), Stathmin (p <0.0001), and the level of chitinase enzyme activity (p =0.0004) were increased in old individuals living in the elderly home (n=20, mean age: 85±8.1) compared to young individuals (n=31, mean age: 30±3.8, figure 2a-d). There was a further increase in serum levels of CRAMP (p =0.0007), EF-1α(p =0.0112) and chitinase enzyme activity (p =0.0215) in hospitalized geriatric patients (n=72, mean age: 73±8.5, figure 2a-d) indicating that increasing serum levels of the marker proteins correlated with aging and aging associated diseases. Multivariate analysis revealed that CRAMP-protein level and chitinase enzyme activity had the best discriminatory power and combination of these 2 markers had the highest statistical power to discriminate the three groups (figure 2e). These markers showed a much better correlation with aging and aging-associated disease compared to interleukin-6 (IL-6) serum level (figure 2f), which is one of the classical biomarkers of human aging . In line with the serum data, IHC revealed increased expression of CRAMP, EF-1α, and Stathmin in post-mortem liver biopsies of aged human (n=5, mean age:76±7.6) compared to younger individuals (n=6, mean age:21±7.7, Suppl. Figure 4, Suppl. Table 2).We next studied the expression of the identified marker proteins in aging human fibroblast (BJ) during in vitro culture. A significant increase of the marker proteins was detectable in pre-senescent cells, as early as 10 passages before terminal senescence of the cultures. The mRNA expression of all 3 orthologs (CRAMP, EF-1α, and Stathmin) and chitinase was up-regulated in pre-senescent cells compared to early passage cells (figure 3a). At protein level, pre-senescent cells showed increased cellular levels of CRAMP, Stathmin, and EF-1α(figure 3b) as well as an increased secretion of CRAMP, Stathmin, EF-1α, and chitinases into the culture medium (figure 3c-e).Several human diseases have been associated with accelerated telomere shortening including cirrhosis, which is the end stage of chronic liver diseases leading to organ failure. mRNA expression of the identified markers were significantly increased in liver biopsies from patients with advanced liver cirrhosis (n=25, mean age: 57±6.5) compared to control biopsies from age matched patients with chronic liver disease before the development of cirrhosis (n=20, mean age: 54±3.2, figure 3f). Similarly, serum levels of the marker proteins and chitinase enzyme activity were up-regulated in cirrhosis patients (n=33, mean age: 58±9.1) compared to non-cirrhotic controls (n=26, mean age: 56±8.4, Figure 3g-k). IHC confirmed up-regulation of CRAMP, EF-1α, and Stathmin on protein level in cirrhosis compared to non-cirrhotic liver (figure 4a).Myelodysplastic syndromes (MDS) are other human diseases, which have been associated with telomere shortening leading to a decline in hematopoietic stem cell function and bone marrow failure . Patients with MDS (n=26, mean age: 69±8.3) showed elevated serum levels of all 4 marker proteins compared to protein levels in unaffected, aged human individuals living in the elderly home (n= 20, mean age: 85±8.1, figure 4b-f).IgA nephropathy is another disease associated with telomere shortening leading toa decline in kidney dusfunction. Patients with IgA nephropathy showed elevated serum lecels of all 4 marker proteins, specially in theⅢand higher grade disease stage ( figure5). IHC show the higher expression of three markers in the disease kidney (figure 5 d-g).Together, this study has identified four secreted proteins that are specifically associated with aging of telomere dysfunctional mice but not with aging of mice with long telomere reserves. The study shows that these secreted proteins are biomarkers of cellular and organismal aging in humans indicating that signaling pathways downstream of telomere dysfunction are activated in human aging and disease.It is possible that secreted proteins induced by telomere dysfunction not only represent biomarkers of human aging but actively impair the function of cells and organs during aging and chronic disease. In agreement with this assumption, telomere dysfunction of stromal cells can have an influence on dedifferentiation of neighboring cancer cells and hematopoietic stem cell function . Interestingly, two of the secreted factors identified in this study (CRAMP and Chitinase) are activated during innate immune responses and inflammatory diseases. The activation of innate immunity is one of hallmarks of human aging linked to cardiovascular disease, neurodegenerative disease, and decreased longevity. The current study indicates that telomere dysfunction can contribute to the activation of innate immunity and could thus help to understand this phenomenon of human aging.In summary, the current study supports the hypothesis that telomere dysfunction influences human aging. The newly identified biomarkers of telomere dysfunction represent powerful biomarkers of human aging and diseases associated with telomere shortening. The study provides a rational basis to begin to explore the influence of the secretome of telomere dysfunctional cells on human aging. Moreover, these newly identified markers can now be tested in clinical studies aiming to improve personalized therapies in the elderly. Such markers could especially be useful when invasive therapies are planned that require regenerative capacity of older individuals (e.g. surgery, chemotherapy, or irradiation).
|
PDF Full Text Request