Advanced age exponentially increases the risk of developing cardiovascular disease, chronic respiratory diseases including COPD and bronchiectasis, and degenerative diseases such as arthritis.(1) Around 100,000 people worldwide die each day due to age-related diseases.(2) Loss of tissue and structural integrity leading to the loss physiological integrity with age, resulting in a progressive decline of homeostasis and reduced capacity to respond to environmental stimuli, is believed to contribute to an incremental risk and severity of these diseases.(3) Several intertwined biological mechanisms including shortened telomeres, epigenetics, inflammation, macromolecular damage, altered metabolism and proteostasis, and reduced stem cells and regeneration have been proposed as key mechanisms driving the loss of physiological integrity in aging and its associated diseases(4). Although most studies have focused on the associations between these processes and chronological age or diseases, less attention has been drawn to their interactions, e.g., whether diseases and age could interact in accelerating aging processes.
Ultra-long lived proteins such as elastin, collagen, and eye lens crystalline have been considered as the Achilles heel of the aging proteome(5) as their damages and losses are not easily repaired. Among them, elastin is unique in providing the characteristics of elasticity, resilience, and deformability of tissues such as the aorta, lung, and skin etc, and its fragmentation and degradation represents an important feature of normal aging.(6) Elastin is a crosslinked polymeric network of tropoelastin monomers catalyzed by lysine oxidase during development. In adult tissues, elastin has an extremely low turnover rate with a half-life of ~ 74 years under normal conditions(7) in contrast to minutes to days for most intracellular proteins(8). In general, adult tissues lack the capability of regenerating functional elastic fibre.(9) These two unique properties imply that an increased turnover of this ultra-long-lived protein in adult tissues could result in irreversible changes to elastin-rich tissues (5). Elastin breakdown products themselves are also known to possess biological activities such as chemotactism, angiogenesis, and inflammatory responses.(10) Monitoring the activity of elastin degradation therefore could be highly relevant to better understanding aging and aging-related diseases.
We have previously investigated the activity of elastin degradation in several age-associated diseases by evaluating the concentrations of circulating total desmosine and isodesmosine (termed cDES), two crosslinking molecules unique to functional elastin(11–14). We found that cDES concentrations were positively associated with chronological age in patients with COPD(15), bronchiectasis(12, 14), abdominal aortic aneurysm (AAA)(13), acute myocardial infarction(16), and in control participants with normal lung function and free of significant diseases.(15) The elevated cDES concentrations were significantly correlated with higher mortality and cardiovascular morbidity or mortality in these conditions.(11–15). Interestingly, the activity of elastin degradation was accelerated by an age-disease interaction in COPD(11), supporting the concept that COPD is a disease of accelerated aging. However, whether a similar age-disease interaction also drives elastin degradation in other age-related diseases is unknown. We set out to investigate whether the age-associated elastin degradation is enhanced in AAA, bronchiectasis, rheumatoid arthritis (RA) and COPD compared to control subjects.
We performed a pooled analysis of data from multiple cohorts comprising individuals with COPD (n = 1332), AAA (n = 507), bronchiectasis (n = 433) or RA (n = 111), and a control group (n = 641). Ethic approvals and informed consent have been obtained. The COPD group was combined from three observational cohorts (the ECLIPSE(17)(ClinicalTrials.gov Identifier: NCT00292552), Nottingham(17), and Scotland cohorts(18)) where individuals with GOLD stage II–IV COPD were recruited. The AAA group included two cohorts (the MA3RS(19)(ClinicalTrials.gov Identifier: NCT02229006) and UKAGS cohorts(13)) where the mean AAA size was 48 mm (standard deviation = 8 mm). The bronchiectasis patients were from the TAYBRIDGE cohort and had a median (IQR) Bronchiectasis Severity Index score of 6(4–10).(14) RA patients meeting the 1987 American College of Rheumatology (ACR) classification criteria for RA (DAS28 score = 4.6 (1.6)) were recruited in Ninewells Hospital, Dundee. With the exception of RA and control subjects, previous history of CV disease, uncontrolled hypertension or hypercholesterolemia and any other inflammatory conditions were not excluded in disease groups. The demographic details of each study group are shown in Supplementary Table S1. The AAA group had a higher percentage of male participants whereas the bronchiectasis and RA groups had higher percentages of females, reflecting the sex differences in the prevalence of those diseases and the fact that only men are screened for AAA in the UKAGS cohort.(20, 21). As expected, the COPD and AAA groups have higher percentages of current and ex-smokers. Our previous studies have shown that sex and smoking history did not affect cDES concentrations in adults(12, 15). Similar to the positive associations to age observed in the COPD and control populations(11), cDES concentrations correlated with age in AAA, bronchiectasis, and RA (r = 0.43–0.48, p < 0.001, figure A-E, Supplementary Table S2). cDES concentrations were measured using the same validated isotope dilution LC-MS/MS method(22) in the same laboratory.
To investigate the effect of age-disease interaction on elastin turnover, we compared the slopes of the regression lines between the four disease groups and the control population using a linear mixed-effects model incorporating subject-specific random effects and taking into account the effects of multiple visits from some individuals (i.e. in the ECLIPSE, Scotland, and MA3RS cohorts). We found that the slope of age to elastin turnover regression in bronchiectasis was greater compared to healthy control subjects (6.27 vs 4.45 ng L− 1 per year, p = 0.02, Figure F). A similar effect of age-disease interaction was found in RA (5.75 vs 3.46 ng L− 1 per year, p = 0.003, figure G). In the AAA group the predicted regression slopes is ~ 5 times greater (16 ng L− 1 per year) when compared to the control population (3.12 ng L− 1 per year; p < 0.001; figure H). The difference remained if those less than 60 years old (defined by the lower limit of 95% interval of the AAA group) in the control group were excluded (p = 0.007 for the interaction). Together with the previous observation in COPD (11), these results strongly suggest that RA, AAA, bronchiectasis, and COPD are diseases of accelerated aging, and that the acceleration is dependent on an age-disease interaction where the differences between the disease groups and control becomes greater as age increases.
Increased cDES concentrations have implications for mortality in COPD(15), bronchiectasis(14), and AAA(13). Notably, the association between cDES concentrations and cardiovascular diseases was particularly strong among these diseases, probably explained by the abundant expression of elastin in the vascular system. Indeed, a significant interaction (greater slope) was also observed in COPD patients with cardiovascular diseases compared to COPD participants without this comorbidity in the ECLIPSE cohort (8.69 ng vs 5.48 ng L− 1 per year, p < 0.037, Figure I). The notion is also supported by the fact that AAA showed the greatest slope among the four diseases studied. Furthermore, cDES concentrations were significantly associated with pulse wave velocity (a measure of arterial stiffness) in patients with COPD(15) or acute myocardial infarction(16) further strengthening its clinical relevance to the cardiovascular system with elastin breakdown mirroring arterial stiffness. These data may provide a potential explanation as to why cardiovascular morbidities and mortality are higher in COPD, RA, and bronchiectasis.
It is unclear how this observed age-disease interactions occur across these diseases caused by different aetiologies. It is plausible that old age increases the expansion of blood cells with somatic mutations (or clonal haematopoiesis)(23), leading to over-activation of immune cells such as neutrophils and macrophages to disease-causing pathogens or reactants(24), subsequently releasing more proteolytic enzymes to the circulation that digest elastin in the vasculature. As such, elevated cDES concentrations signal a harmful inflammatory effect to the cardiovascular system. On the other hand, age associated decline of renal function might also influence the elimination of cDES and cDES containing peptides from the circulation leading to higher cDES concentrations, although our current evidence did not show consistency. In our two previous studies where renal function data are available, we found an association between renal function and cDES concentrations in acute myocardial infarction(16) but not in COPD(15). The reason for this discrepancy is unclear but it is possible that the disease population or patient conditions (e.g., acute vs stable condition) could affect the relationship.
Few proven interventions are available that prevent or minimise elastin degradation in humans. To date, alpha-1 antitrypsin augmentation is the only treatment showing the ability of reducing circulating desmosine levels in a randomised trial although the magnitude of treatment effect was small(25) and it is currently only relevant to alpha-1 antitrypsin deficiency, a rare genetic condition that is associated with emphysema. Next generation systematic anti-proteinase therapy such as DPP1 inhibitors may offer promise especially for bronchiectasis.(26) It would be of great interest to test whether anti-aging therapies such as calorie restriction and caloric restriction mimetics (e.g. metformin) could impact elastin degradation in age-related diseases in human. A recent animal study showing that calorie restriction was able to reduce arterial aging by reducing proinflammation associated elastin degradation supports this concept.(27)
We acknowledge several limitations in this study. First, we were not able to estimate the contribution of renal function to the observed age-disease interaction due to the lack of data availability. Second, the comorbidity data were not always available and therefore the influence of the comorbidity cannot be always excluded. Third, we did not include the acute myocardial infarction study(16) in the current study as the data were not available.
In summary, we conclude that accelerated elastin degradation by age-disease interaction is a common feature in age-related diseases.