For probably the first time in history, persons over the age of 65 outnumber children under 5 years of age . With increasing life expectancy however, there has occurred an increase in infectious disease severity, neoplasia, and autoimmune disease as a consequence of immune senescence in the elderly [2, 3]. This poses challenges to public health and medical care in the elderly for vaccination against infectious diseases and treatment of chronic diseases, respectively. For example, influenza and pneumonia are among the top ten causes of death in the elderly, yet influenza vaccination is only 17 – 53% effective in elderly adults compared to 70 – 90% efficacy in younger adults . Defining mechanisms of immune senescence is expected to help generate intervention therapies and establish predictive markers to identify those individuals requiring supportive strategies such as vaccine boosting, for example.
Nonhuman primates provide a vital link for translating basic science research to applications in humans for improving well-being in the elderly [7, 12]. Advantages to using nonhuman primates include genetic homology, physiology, behavior, and responses to infections and diseases that are shared with humans. In addition, nonhuman primates are outbred, and outdoor-housed animals accumulate environmental exposures similar to humans [9, 13]. Clinical, medical, and pedigree information is available for captive nonhuman primates used in biomedical studies, as well. Medications and compliance of their use by elderly humans may affect interpretation of results from studies on natural biological aging, so studies using nonhuman primates may control for these effects. Nonhuman primates also serve as analogous models of humans with similar and often identical susceptibility to infectious agents. This provides another advantage because nonhuman primates can be experimentally challenged with infectious disease agents to validate vaccine or drug efficacy which usually cannot ethically be accomplished in humans. Further validating this nonhuman primate model in gerontology are reports that dietary restriction extending median and maximum lifespan in a wide range of animals such as worms, fruit flies, and mice, also extends life-span or delays onset of age-related chronic diseases in nonhuman primates, thus supporting translation to humans [14–17].
Chronic low-grade inflammation is associated with aging but questions remain whether inflammation is causal, coincidental, or consequential to aging, and which biomarkers of inflammation are relevant to studies on immune senescence [4, 6, 18, 19]. Rhesus macaques are among the most commonly-used nonhuman primates in biomedical research and provide applicable models that simulate human physiology and inflammation related to aging [7, 9, 12, 20]. A goal of this study therefore, was to correlate immunological parameters from blood samples of outdoor-housed rhesus macaques to develop a model of immune-senescence that can be used to study mechanisms relating inflammation and aging.
The rhesus macaques used in this study were housed outdoors and ranged from 2.5 to 24 years of age that is roughly equivalent to 8 – 77 years of age in humans. A cross-sectional experimental design was applied to correlate characteristics of blood, and especially cytokine/chemokine concentrations over age and compare means between older and younger animals to define significant characteristics associated with aging. Neonatal and infant animals were excluded so that developmental characteristics would not confound the correlations that often shift with maturation. In addition, exceptionally long-lived rhesus macaques aged 25 years or more (approximately equivalent to 80 years of age in humans) were not included because frail older animals will have died and characteristic trends of overall aging were reported to deviate or change in exceptionally longer-living older animals . Excluding exceptionally longer-lived animals also helps distinguish correlates related to primary changes of aging from secondary changes affected by diseases of aging. The preponderance of females to males at an approximate ratio of 3:1 reflects the population demographics of the outdoor breeding colonies at the TNPRC and gender differences were only rarely observed in these studies.
Initial studies on blood chemistry and hematology performed on outdoor-housed animals were similar to results previously published on indoor-housed rhesus macaques. An extensive study and update by Smucny et al. describing characteristics of blood chemistry and hematology during aging in captive rhesus macaques identified significant declines in albumin, albumin/globulin, creatinine, MCV and MCH along with significant increases in alkaline/phosphatase, BUN, BUN/creatinine, Hct, Hgb, and RBC numbers [22, 23]. In this current study, significant inverse correlations with age were reported for albumin, creatinine, and aspartate transaminase (AST). AST was not measured in the Smucny report and triglycerides were not measured in the study reported here. There also was no significant correlation between erythrocyte numbers and age in the outdoor-housed rhesus macaques of this study, and although mean Hct and Hgb levels were lower for females than males, there were no significant differences in correlation coefficients between genders over age.
No significant changes in circulating levels of WBC were reported in the Smucny studies over age whereas significant inverse correlations over age were observed for WBC in the studies reported here and as also reported for humans previously [22–24]. Results reported on the outdoor-housed macaques also demonstrated declines in circulating levels of lymphocytes, monocytes, neutrophils, eosinophils, and basophils over age. The decline in blood lymphocyte levels overall was reflected primarily from decreasing levels of B cells with age, rather than T cells or NK cells. Several reports suggested a consensus that overall T cell levels in blood do not change with age, as also observed here, and while variable results were reported in humans and mice regarding levels of NK and B cells, most of these studies described declining functional capabilities of these cells with aging [25–28].
Overall numbers of circulating CD4+ T cells and CD8+ T cells did not significantly correlate with age in this group of outdoor-housed macaques. Within the T cell populations, however, there was a significant direct correlation between CD8+ effector memory T cells and significant inverse correlations between CD4+ and CD8+ naïve T cell populations with age. These findings are similar to reports by others [29–33], including a study on indoor-housed rhesus macaques reported by Cicin-Sain and colleagues comparing mean levels between younger (aged 6–9 years old) and older (aged 18 – 24 years old) rhesus macaques. In addition, we observed a significant correlation between increasing levels of DP (CD4+CD8+) T cells in blood with increasing age, and numbers of DP effector memory cells statistically significantly correlated with age while numbers of DP naïve and central memory cells did not significantly correlate with age.
Identifying biomarkers of aging may be important for predicting vaccine failure or susceptibility to infectious disease pathogens, cancer or other diseases that are especially problematic in the elderly. Chronic low-grade inflammation is considered a hallmark of aging, and multiplex cytokine analysis was applied in attempt to identify a panel or group of systemic cytokines that together significantly correlate with aging. Such a panel of cytokines is expected to be more consistent than an individual cytokine/chemokine level for selecting individuals with relatively higher and lower levels of inflamm-aging into research studies. The ability to select groups of animals with high and low inflammation status will be important to test hypotheses and mechanisms relating cause and effect between inflammation status and aging. This would also open the door to developing strategies to restore immune competence or delay onset of immune senescence.
This study is among the first to apply multiplex cytokine analyses in rhesus macaques for developing a model of immune senescence. The outdoor cohort of macaques exhibited increasing levels of several pro-inflammatory cytokines with age, including IFNγ, IL1β, IL6, IL12, IL15, MIF, sCD40L, and TNFα with age, similar to observations by others [18, 34–36]. Of interest was the observation that while circulating levels of IFNγ, IL6, IL15, MIF, and CD40L correlated directly with age, t test comparisons between mean levels in the older group of macaques (aged 18 – 24 years) were not significantly higher than those in the younger group of macaques (aged 2 – 9 years), in part due to higher variability exhibited by wider standard deviation values. Such findings may be especially prominent in studies of outbred populations such as humans and nonhuman primates than perhaps would be expected to occur inbred strains of mice. The higher variation in older groups might also explain why others reported no significant differences between younger humans (less than 45 years of age) and older humans (65 years or more) when comparing mean circulating levels of cytokines such as IFNγ, IL6, IL12, or TNFα after multiplex testing and applying t test analyses .
Individual cytokines behave in complex, redundant, and pleiotropic mechanisms making it difficult to relate any single pro-inflammatory cytokine to biological aging. In addition, the increasing standard deviations in mean concentrations of several circulating pro-inflammatory cytokines suggested that not all animals exhibited systemic inflamm-aging at the same rate as chronological aging. A goal of this study was to establish a basis for selecting animals exhibiting relatively higher and lower levels of cytokines/chemokines associated with inflamm-aging so that subsequent studies can be planned to test how inflammation affects susceptibility and development of chronic diseases in the elderly. In attempt to define a group of cytokines/chemokines that reflect immune senescence, Pearson pairwise correlation analysis was applied to each of the cytokines assayed in the multiplex testing platform. A “concordant” set of cytokines/chemokines was identified whose levels correlated with aging and that also exhibited significant pairwise correlations with each of the other factors of this group. These included the pro-inflammatory cytokines IFNγ, IL1β, IL6, IL12, IL15, MCP1, MIPα, and TNFα. Of interest was the observation that two anti-inflammatory cytokines, IL4 and IL1ra, also exhibited significant correlations with each of the other concordant cytokines in this group. The increased levels of circulating IL4 and IL1ra levels correlating with age may indicate repair responses to inflammation or may reflect dysregulation of cytokines and chemokines purported to occur during immune-senescence [36, 38]. To ascertain that there were no outlier cytokine concentrations that could confound interpretation of these results, mean rankings rather than cytokine concentrations were analyzed and found to significantly correlate directly with age whereas mean rankings of the remaining cytokines as a group did not correlate with aging. Furthermore, the panel of 10 concordant cytokines exhibited a more robust correlation with age than did a smaller group of number of cytokines, IFNγ, IL1β, IL6, IL12, and TNFα, that are routinely associated with inflamm-aging. The use of mean ranks compressed the dynamic range of analyses to generate more conservative correlation coefficients, but demonstrated that the mean rankings of cytokines containing both pro- and anti-inflammatory activities exhibited greater statistical significance than the mean rankings of the five pro-inflammatory cytokines commonly associated with aging. These results provide a model to compare animals exhibiting relatively higher versus lower levels of these cytokines and chemokines to relate chronological aging with biological aging as well as address impact on susceptibility to chronic diseases typically associated with aging. Also of interest was the observation that a few of the younger animals exhibited relatively higher inflammatory cytokine rankings, and longitudinal studies could be performed to determine if these higher cytokine/chemokine-ranking younger animals were responding to recent infections or are pre-disposed to biologically age faster than their lower ranking cohorts.
Risk factors associated with immune senescence have included chronic infections with cytomegalovirus (CMV) and obesity. We compared body weights of the older 18 – 24 year-olds with high and low ranking circulating chemokine/cytokine levels (above or below 50, respectively; Figure 4, panel A) among females (8.99 kg ± 0.46 vs 8.02 kg ± 0.29, respectively) and males (14.35 kg ± 1.46 vs 12.35 ± 1.71, respectively) and found no statistically significant differences within each gender. The rhesus macaques used in this study are housed outdoors and are thus less sedentary than indoor-housed animals which could explain this observation. In addition, animals over age 24 were not used in this study to preclude changes that may be associated with chronic diseases of aging rather than aging itself. Chronic CMV infection is believed to promote the higher levels of effector-memory T cells that affect the overall immunological cell population profiles that shift with aging. Based on a summary of the second CMV and immunosenescence workshop, many questions remain about the impact of CMV infection on the aging immune system and whether treatment to clear CMV, for example, would delay or reverse immune senescence . Monkeys are not routinely tested serologically for CMV at the Tulane National Primate Research Center, but a limited study of the SPF colony at the Tulane National Primate Research Center several years ago determined that approximately 20% of the animals were CMV seronegative and that these were all among the younger group of rhesus macaques. All of the animals in the older group were thus highly likely to be CMV seropositive suggesting that CMV infection alone was not the cause for the varied inflammation status observed in the older rhesus macaques. Further studies, however, are needed to corroborate this.
The results of this study confirm and extend the benefits afforded by using nonhuman primates to study aspects of aging related to systemic pro- and anti-inflammatory cytokines/chemokines. From the literature, there seems to be consensus that increased levels of circulating inflammatory cytokines/chemokines are indicators of increased biological aging and risk for frailty in the elderly [12, 18, 38, 39]. It is unclear, however, whether age-correlating changes in these cytokine/chemokine levels represent biomarkers of aging, contribute to chronic diseases of aging, are produced as a consequence of developing diseases of aging, or combination of these possibilities. Continued studies using this nonhuman primate model will allow us to test whether the higher ranking animals putatively exhibiting “inflamm-aging” are more susceptible than the relatively lower ranking animals to experimental infectious disease or vaccine failure. Such a model can then be used to discern between primary changes related to aging from secondary effects of diseases of aging that would also affect inflammation.