One key goal of basic aging research is the development of reliable assays of both current and future health. These assays could dramatically accelerate progress toward developing health-extending interventions by obviating the need for full lifespan studies, especially if they were informative relatively early in life. One potential approach is the assessment of physiological resilience, defined as the ability to recover from an adverse event. Here, using CB6F1 mice, we evaluated four potential resilience assays, each quantifying recovery from a physiological challenge with clear relevance to humans. The challenges were: (1) anesthesia recovery, (2) restoration of hemoglobin levels after a blood draw, (3) speed of wound healing, and (4) survival after pathogen exposure. We evaluated how each changed with age and with interventions known to extend health in males only (17α-estradiol) or both sexes (calorie restriction). We found that three of the four (recovery from anesthesia, blood draw, and pathogen exposure) showed significant and expected age effects, but wound healing did not. None of the three age-sensitive assays responded to the health-extending interventions in the way we expected, and for some assays, including anesthesia response, interventions actually worsened outcomes.
Possible explanations are: (1) our interventions were too brief, (2) the ages we evaluated were too young, (3) our assays did not capture important features of organismal resilience, or (4) organismal resilience is not as clearly related to current or future health as hypothesized. Future studies are needed to determine which of these interpretations is valid and to determine whether other resilience metrics may be more informative about current and future health.
Experimental interventions
We evaluated the response of our four resilience assays to two previously described life- and health-extending treatments, dietary 17α-estradiol (EST) supplementation and calorie restriction (CR).
17α-estradiol dietary intervention.
Twelve months old mice underwent baseline assessment of anesthesia and blood draw recovery just prior to randomization into control or intervention arms of the study. A control diet, TestDiet 58YP (66.6% carbohydrate, 20.4% protein, 13.0% fat) was supplemented with 17α-estradiol at 14.4 ppm to make the intervention diet. This is the same diet previously shown to extend male only life and health in UM Het3 mice [27, 32]. Diets were prepared by TestDiet, a division of Purina Mills (Richmond, IN). Both control and treatment groups were provided food and water ad libitum throughout the duration of the study. Resilience assays were repeated after 3 months and again after 8 months (mouse ages 15 and 20 months, respectively) after the introduction of the intervention diet.
Discussion
To summarize, two of our four resilience assays (anesthesia recovery, pathogen challenge) showed robust age effects, one (recovery from blood draw) showed a statistically marginal trend in the direction of declining function with age, and the fourth (wound healing of an ear punch) showed no hint of an age impact. Our EST supplementation experiment did not show a general improvement in resilience in male mice only as our hypothesis predicted, if anything EST males performed worse on the resilience assays, but like previous studies, males were the only sex affected by the drug in several parameters. EST supplement did not affect male recovery from anesthesia or from blood draw, but it did lower baseline hemoglobin levels, suggesting that it may be having some negative physiological effects. It has been established previously, that 17β-estradiol can inhibit erythropoiesis [41], and our results suggest 17α-estradiol may also show similar phenotypes. In addition, EST had marginal impacts on male movement in the open field but in the opposite direction than expected if EST improved male health. Similar to previous reports [29], we found that EST-treated males, but not females, lost weight—largely body fat—during the experiment and never recovered to baseline, whereas control males continued to gain weight as they aged. There was no impact of EST on anesthesia recovery in females, but males at 8 months recovered more slowly, opposite to what our hypothesis predicted. EST also had no impact on recovery from a blood draw in females. However, males had a significant decrease in baseline hemoglobin levels on EST–opposite to our hypothesis. One behavioral test, quality of nest-building did show a positive effect of EST in males but not females, suggesting that this simple assay may be worth investigating further, as this is an indicator of overall mental and physical health of the animals [34, 42]. The other behavioral assays were not affected by EST supplementation in either sex. In our CR experiment, anesthesia recovery was worse, not better, in CR-treated mice of both sexes, again suggesting a negative physiological effect of the lifespan extending intervention.
All of our resilience assays except wound healing exhibited either a statistically significant, or trend, toward declining with age as expected. The age effect would no doubt have been greater if we had employed a wider age range, but we felt it to be unlikely that resilience assays done prior to 12 months of age would be informative about later life health and survival. Recent resilience work in mice found a similar age effect for anesthesia response [17].
The wound healing assay surprisingly did not show any hint of an age effect, even though age has been reported to slow wound healing in laboratory mice (and rats) previously [22, 43]. However, typically the animals in those experiments were considerably younger than our youngest animals, and wounds were full-thickness wounds in the interscapular area rather than ear punch wounds. For instance, Cohen and colleagues [43] found that full-thickness interscapular area wounds healed more quickly in young (6 mo) male C57BL/6 mice than in mature (15 mo) or old (26–27 mo) mice. We should note, however, that young rodents do not always show faster healing than older ones, in some species such as longer-lived Peromyscusmice, old mice heal faster [44] and in C57BL/6 and BALB/c mice young mice (1–2 months) have worse wound healing abilities than middle aged (12 month) mice [45]. We piloted the use of full-thickness interscapular wounds in our studies, finding that they worked well for C57BL/6 mice which could not reach the wounds with their paws to scratch at them, whereas our more supple CB6F1 mice could. CB6F1 mice were observed by investigators to regularly chew or scratch at wounds and were readily able to dismantle or escape splints. Thus, it should be noted that mouse genotype can affect the feasibility of doing certain assays. Ear punch wounds were particularly attractive alternative as a model of wound closing as they are often done routinely to individually identify animals in a cage. Ear punches do close partially, even in older animals, and the rate of ear punch closure in old mice has been associated with overall health in the animals [46]. However, to be most useful for measurement of age-related changes, and to clearly distinguish between wound contraction and wound closure, more sophisticated approaches such as the use of splints or scaffolds to stabilize wound area may be necessary [47], and we were using simple measures of size of the wounds without looking at histological changes. Potentially we would find similar rates of wound closure in middle and old age mice, but differences in the “new” tissues structure. Another disadvantage of wound healing protocols as possible markers of biological aging in individuals are that they are difficult to do multiple times longitudinally on the same animal.