Urinary markers of oxidative stress respond to infection and late-life in wild chimpanzees

Oxidative stress (OS) plays a marked role in aging and results from a variety of stressors, making it a powerful measure of health and a way to examine costs associated with life history investments within and across species. However, few urinary OS markers have been examined under field conditions, particularly in primates, and their utility to non-invasively monitor the costs of acute stressors versus the long-term damage associated with aging is poorly understood. In this study, we examined variation in 5 urinary markers of oxidative damage and protection under 5 validation paradigms for 37 wild, chimpanzees living in the Kibale National Park, Uganda. We used 924 urine samples to examine responses to acute immune challenge (respiratory illness or severe wounding), as well as mixed-longitudinal and intra-individual variation with age. DNA damage (8-OHdG) correlated positively with all other markers of damage (F-isoprostanes, MDA-TBARS, and neopterin) but did not correlate with protection (total antioxidant capacity). Within individuals, all markers of damage responded to at least one if not both types of acute infection. While OS is expected to increase with age, this was not generally true in chimpanzees. However, significant changes in oxidative damage were detected within past-prime individuals and those close to death. Our results indicate that OS can be measured using field-collected urine and integrates short- and long-term aspects of health. They further suggest that more data are needed from long-lived, wild animals to illuminate if common age-related increases in inflammation and OS damage are typical or recently aberrant in humans.


examples.
This statement is required for submission and will appear in the published article if the submission is accepted. Please make sure it is accurate.
Unfunded studies Enter: The author(s) received no specific funding for this work. All study methods were non-invasive and observers maintain a distance of 5 m from chimpanzees at all times. Yes -all data are fully available without restriction Background: For decades, oxidative stress has been a focal point of research on aging in humans and laboratory animals. More recently, it has been examined as a mechanism of trade-offs by which the wear-and-tear of life translates into biological decline in capture-sampled vertebrates. Markers of oxidative stress are commonly assayed in blood and sera, however many studies of wild animals are restricted to non-invasive sampling. As gerontology and evolutionary ecology begin to integrate, there is a need for reliable tools to evaluate oxidative damage and protection non-invasively in the wild, particularly as wild non-human primates and great apes provide valuable comparative data for human aging.

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
Our study: We performed a validation analysis of 5 urinary biomarkers of oxidative damage and protection in field collected urine from a community of wild chimpanzees in The Kibale National Forest, Uganda. We capitalized on the chimpanzee community's long-term health records, a rich collection of urine samples, and natural experiments to evaluate 1) the acute effects of infection and 2) the potential cumulative effect of age on urinary oxidative status. As a proof-of-concept, we evaluated acute responses in oxidative markers to a widespread respiratory epidemic and also to instances of severe open-wound injuries. We then further evaluated whether markers of damage increased with age in a mixed longitudinal sample and also within individuals approaching death.

Findings:
We found that all markers of damage responded to at least one validation treatment, and that our marker of DNA damage positively correlated with all other damage markers, signaling its utility. Importantly, we find that different types of damage resulted from different kinds of infection (bacterial vs. viral) and in different time courses. This highlights the complexity and dynamism of oxidative status and the need for evaluating it with multiple key markers. Counter to predictions, we found no increase in oxidative damage with age in our mixed-longitudinal sample, though increased damage did appear in individuals as they approached death. Further, we found no overt signal of immunosenescence in chimpanzees, though again, an increase in lipid damage and inflammation among senescent individuals, suggesting that human-like patterns of inflammaging might occur in extreme cases.
Significance and novelty: Our study highlights markers of DNA damage, lipid peroxidation, and inflammation as useful tools to non-invasively monitor the oxidative status of wild primates. Our results further warrant future research on great apes and non-Western and post-industrialized human populations to evaluate whether inflammaging is a recent and atypical development in humans.
We expect this work to be of broad interest to researchers in disciplines including ecology, gerontology, and public health. We, the authors, thank you for your consideration of this work.
2 examine responses to acute immune challenge (respiratory illness or severe wounding), 24 as well as mixed-longitudinal and intra-individual variation with age. DNA damage (8-25 OHdG) correlated positively with all other markers of damage (F-isoprostanes, MDA-26 TBARS, and neopterin). Within individuals, all markers of damage responded to at least 27 one if not both types of acute infection. While OS is expected to increase with age, this 28 was not generally true in chimpanzees. However, significant changes in oxidative 29 damage were detected within a few years of death among aged individuals. Our results 30 indicate that OS can be measured using field-collected urine and integrates short-and 31 long-term aspects of health. They further suggest that more data are needed from long-32 7 Protection from oxidative damage can also be measured non-invasively, as urine 138 contains viable antioxidants from both endogenous and dietary origin. Total antioxidant 139 capacity (TAC) is a measure of oxidative resistance indicated by the ability of a medium 140 to prevent a redox reaction between controlled substances  and 141 has been validated to negatively correlate with isoprostane levels . 142 Endogenous antioxidants, in particular, often vary with age in a complex manner, as their 143 basal expression increases but their inducibility of expression decreases with age in 144 humans and rodents .  Table 1). This period was 284 bookended by severe peaks in respiratory signs. We then identified 7-month 285 periods "before" and "after" the prolonged outbreak. Given a single week when 286 several members showed severe respiratory signs 2 months prior to the prolonged 287 outbreak, we excluded these 2 months and restricted our control sample to the 288 window 2-7 months before the outbreak. All individuals were sampled for at least 289 two periods for any given biomarker. For specific sampling per biomarker, see 290 Table S1. 291 292 2) Before, during, and after serious injury: As a further proof-of-concept that acute 293 challenges lead to OS, we examined OS biomarker responses to traumatic injury 294 in case studies of three individuals. We selected cases that were severe enough 295 that they posed a strong risk of bacterial infection. Two chimpanzees, GG (14 yo 296 hand; these resulted in significantly impaired movement and took >2 weeks to 305 heal. All subjects were sampled over the 3 months before the injury, during the 306 period between injury and healing (by visual assessment for bleeding, 307 inflammation, infection, and scabbing), and for 3 months after the injury healed. 308 For sampling date ranges by individual and biomarker, see Table S2.  Table S3. response, hour of sample collection as a fixed effect, and individual as a random intercept 336 with hour of collection as a random slope. To minimize the influence of potential outliers, 337 these models were applied to samples collected during baseline conditions, i.e. not 338 collected during or after a respiratory epidemic or injury, and not from the four 339 individuals whose deaths were the focus of our analyses above. Because MDA-TBARS, 340 neopterin, and TAC varied significantly with time of day (Table S4) Table 2). The three subjects showed no significant variation in 8-OHdG, 412 or TAC levels in response to injury ( Figure 3A, 3C, 3E). 413 Oneindividual (GG, snared, Figure 3) exhibited an unusual pattern for all of the 414 biomarkers examined. While GG's isoprostanes increased markedly in response to injury, 415 unlike the other individuals, they continued to increase after the wound was visibly 416 healed. On the other hand, while her levels of neopterin were similarly high during the 417 injury phase as the other two, and her post-healing levels were similarly low, she had 418 already exhibited highly elevated neopterin and MDA-TBARS, as well as 8-OHdG prior 419 to her injury. It is possible that GG was suffering an unknown condition despite having 420 exhibited no other visible clinical signs over the pre-injury period. 421  0.11 ± 0.13 0.04 ± 0.15 -0.07 ± 0.14 0.53 ± 0.13 *** -0.31 ± 0.14 * † p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001. Sampling date ranges per individual in Table S2. In this study, we examined a set of urinary biomarkers to determine their utility 463 for assessing oxidative stress in future studies of health and aging in wild chimpanzees 464 (Table 5). Redox dynamics in the body are highly complex and, accordingly, our results 465 were not always straightforward. We found that DNA damage (urinary 8-OHdG    , 2017). Surprisingly, however, neither isoprostanes nor 504 MDA-TBARS correlated with neopterin. Generally, the modest associations across 505 markers suggest that they each provide independent information on oxidative status. As 506 24 our experiments suggest, this may be due to variable pathways of oxidative damage 507  or to differences in time scales of response. 508 We predicted that total antioxidant capacity (TAC) would be negatively 509 associated with markers of damage because it is this imbalance that characterizes 510 oxidative "stress". As damage-causing radicals increase, more antioxidants should be 511 depleted, as during activities such as exercise . In fact, 512 however, we found that TAC correlated positively with lipid damage (isoprostanes) and 513 inflammation (neopterin). One reason that TAC may increase alongside markers of 514 damage is that pro-oxidants activate Nrf2 transcription factor, which increases the following infection, after which concentrations declined but remained elevated from 539 baseline for several weeks . The fact that neopterin 540 appears to spike briefly, even when the infectious agent persists, means that it can be 541 difficult to capture this signal. At Kanyawara, we did not observe a significant increase in 542 neopterin during the course of a viral respiratory outbreak, despite relatively high 543 sampling intensity. However, we did observe increased levels in the period after the 544 respiratory outbreak. By contrast, neopterin levels increased acutely when chimpanzees 545 had severe open wounds, a time of potentially high infection risk, and returned to levels 546 that were close to baseline after injuries were healed. 547 Unlike the short-term and acute respiratory infections monitored in Taï  548 chimpanzees and rhesus macaques, the respiratory epidemic in the Kanyawara 549 chimpanzee community lasted for 4 months, with three waves of respiratory signs 550 suggesting possible reinfection (Scully et al., 2018). Indeed, 47 of 53 community 551 26 members were seen coughing and 31 had productive coughs during multiple months of 552 2013, and deaths occurred over a span of three months. While it is surprising that our 553 opportunistic sampling procedure failed to capture the expected spikes in neopterin 554 during the outbreak, the elevation in levels after the outbreak suggest that such prolonged 555 (or repeated) illness exerts somatic costs that far outlive the active infection. Whereas 556 abrupt increases in neopterin can be useful to identify infections (in cases that might not 557 be as obvious as respiratory illnesses), it is these lasting inflammatory effects, related to 558 increased oxidative damage, that are predicted to be of particular relevance to longevity. patients with muscular dystrophy  and between certain age 571 categories in a large sample of men and women (Gan et al., 2018). However, it did not 572 increase with age in silica miners, and instead was simply higher in miners with vs. 573 without silicosis (Pilger et al., 2000). 574 27 In contrast to our mixed-longitudinal analysis, isoprostanes and neopterin did 575 increase among very old chimpanzees in their years prior to their deaths. A possible 576 reason for the difference in results between our mixed-longitudinal and within-individual 577 age-based analyses is selection bias, as chimpanzee populations have higher variation in 578 mortality than human populations typically do ( The signal of immunosenescence in chimpanzees, characterized by increased 587 inflammation with age, appears weaker than it does in humans where immunosenescence 588 is prominent in both longitudinal and cross-sectional studies. Indeed, in humans increased 589 inflammatory activity is a core component of aging that is visible cross-sectionally in 590 inflammatory markers (neopterin), to monitor acute oxidative stress and its potential role 635 in age-related declines in health and mortality in wild primates. We advise measuring 636 both DNA and lipid markers simultaneously, as they may indicate different pathways of 637 oxidative damage resulting from different pathogenic challenges and age. 638 Although a larger sample is required to authoritatively delineate age-related 639 patterns of OS and immunosenescence in chimpanzees, we found no evidence that 640 oxidative stress generally accumulates or that immune regulation generally declines with 641 age in this species. Oxidative damage and inflammation do, however, appear to be 642 30 associated with healthspan, as they increase in particularly old individuals that are 643 approaching death. Such findings broadly warrant further research on non-WEIRD 644 humans and apes (Henrich et al., 2010) to identify if the common occurrence of 645 inflammaging in humans is evolutionarily aberrant. Lastly, the limitation of urinary 8-646 OHdG as a marker of both DNA damage and repair emphasizes the value of both current 647 invasive sampling to characterize age-related changes in DNA damage, and future 648 development of assays for DNA repair processes in non-invasive samples. 649