Here, we report the effects of lifelong CR on cognitive function in male F344×BN rats across a wide range of ages, including very advanced age (38 months of age). Performance was assessed using two tasks that are sensitive to detecting age-related cognitive decline: object recognition and MWM spatial discrimination. In accordance with the literature, we demonstrate a dissociation across tasks such that all ages performed equally in object recognition, whereas there was progressive age-related impairment on the memory component of the spatial discrimination procedure. In addition, we demonstrate that CR may improve selected parameters of learning (proximity and latency to reach the platform) in the spatial discrimination version of the MWM, although not affecting others such as distance, which is a more sensitive measure in that it is not confounded by performance (swimming speed) in this task. CR also failed to affect memory on this task (as measured in the probe trials) and perhaps actually impaired memory in the object recognition task. Several parameters may explain these effects and are discussed in detail later.
Consistent with our findings, there are published reports suggesting that performance in the object recognition task is not disrupted with age in rodents. For example, two separate research groups (31
) have compared 6- and 22- to 24-month-old male rats across two different strains: F344 and Wistar rats. Similar procedures were employed in each study such that following a habituation phase, rats were exposed to five salient objects in the testing arena. On subsequent trials, rats experienced both modification of the spatial location and novelty of those objects. Results were consistent across both studies such that older rats, when exposed to a “novel” object, spent the same amount of time exploring that object as did the young rats. However, when confronted with the spatial displacement of a familiar object, old rats explored all objects less, whereas young rats increased exploration of that displaced object.
To our knowledge, there is only one other study currently published in the literature using F344×BN rats in the object recognition procedure (19
), and, in contrast to our findings, these investigators report an age-related decline (between 4- and 32-month-old rats). Procedurally, the task is run in exactly the same way with the exception of a 10-hour retention test (24 hours in the current study) and exploration times in the sample and test phases of only 3 minutes compared with 5 minutes in our study. Future studies need to parametrically vary retention test and time allowed for exploring objects across a wider range of ages in order to more fully characterize object recognition memory in the F344×BN rat.
There are no data in the literature regarding the effects of CR on object recognition performance in any strain of rat. In our study, CR rats actually performed more poorly across all ages relative to their AL-fed controls. This was most likely explained by the CR group’s overall increase in physical activity as measured during the habituation phase and by the increased time spent exploring both objects during both the sample and the test phases relative to the AL group. Therefore, rats never habituated to the testing environment and were most likely in a heightened motivational state due to their fasted status.
This motivational status most likely played a similar role with regard to performance of the learning component of the MWM spatial discrimination task. As has been previously reported by us and others (13
), we observed a progressive age-related increase in latency to reach the platform, which was reversed with CR. The reduced latency to reach the platform in both aged and CR rats may best be explained by the overall increase in swim speed in CR rats. Interestingly, when using proximity as a measure, which is influenced less by swimming speed, age effects were mitigated and CR rats actually performed better than rats in the AL group. The spatial discrimination procedures also allow for the evaluation of acquisition of a spatial search strategy and memory or the ability to retain learned information over time. During the probe trial, we observed a decrease in the memory index, quadrant time, and proximity across age at both the immediate and the 24-hour time points, that was unaffected by CR.
There are a handful of studies that examine the effects of CR on cognitive performance using the MWM in aged rats, with differing results. For example, Markowska and colleagues (12
) have demonstrated that CR led to the improvement of performance in the spatial discrimination version of the MWM using the F344×BN strain (13
), whereas no effect was noted in the parent BN strain or was previously reported in the F344 strain (12
). They conclude that genetic variation may account for these differences although these investigators failed to assess swimming speed during the actual testing. Thus, this difference may perhaps be due to genetic variability in baseline level of locomotor activity or vulnerability to locomotor activity manipulation by CR across these strains. Future studies are merited to make these distinctions.
A final caveat concerns the nutritional status of the rats at the time of testing. More specifically, it is unclear whether the effects observed using the MWM are dependent on the acute or long-term effects of CR and the timing of feeding (during the light cycle) and how this status contributes to increased physical activity. For example, there is clear evidence in both rats and mice that the AL-feeding cycle acrophase occurs in the middle of the dark cycle (21
). Therefore, on the one hand, both CR and AL rats could potentially be in a relatively equally short-term fasted state during behavioral tests. On the other hand, it is also the case that CR-maintained rodents fed during the light phase exhibit patterns of motor activity, body temperature, and metabolic energy expenditure that are independent of the light cycle and instead dependent on the feeding time. The result is a full or partial phase shift in motor activity from that of AL-fed rats (21
). In particular, CR rats exhibit an anticipatory increase in physical activity, temperature, and metabolism before feeding time. In the context of the current studies, we eliminated this confound by feeding our CR rats just prior to the onset of the dark cycle, and our results are very similar to those obtained by Markowska and Savonenko (13
) using phase-shifted F344×BN rats.
In light of these findings, there has still never been a study that examines the dissociation between the effects of chronic, lifelong food deprivation versus short-term food deprivation on learning and cognition. However, two studies in mice and rats have shown that nutritional supplementation with glucose just prior to cognitive testing facilitates learning (39
). For example in rats, Yanai and colleagues (41
) demonstrated impaired cognition in rats receiving CR on performance in the MWM which was reversed with administration of an intraperitoneal glucose injection 30 minutes prior to testing where no other detrimental effects are observed. Still, whether this is an effect of acute glucose administration on learning and memory or whether this is akin to “satiation” in the context of lifelong CR is still an empirical question. In addition, neither of these studies investigated whether an equalization of physical activity contributed to these effects. Therefore, consideration of these influences is critical to the study of learning and memory with regard to lifelong CR and aging.
An additional finding of this study is that CR delays declining physical function in advanced aging (38-month-old rats) even in the absence of a ubiquitous effect on cognitive function. This was reflected in the increase in grip strength and overall increase in a time-limited assessment of movement in all CR groups. Notably, there was a decline in strength and movement across ages regardless of diet. Somewhat controversial is the correction of grip strength performance to whole body weight when there are such large differences between AL- and CR-fed rats not only in overall weight but also in terms of body composition (fat vs lean components). In our experiment, we did not directly measure total fat and lean compartments, so that a correction of grip strength based on muscle mass was not possible, although CR rats surely have a reduced lean mass relative to AL-fed rats based on previous studies. However, several studies in humans suggest that declining strength with age is not merely a result of a reduction in lean skeletal muscle mass but may be better predicted by fat mass and/or fat to lean ratio (42
). Therefore, we use whole body weight as our correction factor. Future studies will include the use of imaging technologies (such as time domain nuclear magnetic resonance) to further investigate the combined and individual contribution of these compartments to performance.
We and others (30
) have previously demonstrated that across a more narrow range of ages, physical function is increased with lifelong CR. This is consistent with our current findings but we further extend this into very advanced aging. What is particularly interesting is that this may be accounted for by the fact that across all ages, again there was an increase in physical activity and/or function in the CR groups. Interestingly, Minor and colleagues (47
) have demonstrated a similar dissociation using mice as participants. Therefore, CR may differentially affect the measurement of physical ability and cognitive function as each is currently and conventionally tested, due to increases in physical activity. In addition, these studies demonstrate that although the field focuses on the importance of “biologic mechanisms” that contribute to CR effects on a variety of functional outcomes, “behavioral mechanisms” play just as likely a role in determining health effects outcomes and may have potentially greater clinical application. Finally, we hope this study raises awareness to the complications of studying learning and memory in the context of CR with regard to the genetic, motivational, and nutritional status of rats used as participants and expands the age range across which the effects of lifelong CR are studied and understood.