The results of the present study demonstrated that elderly rats exhibited diminished spontaneous motor activity, increased anxiety-like behavior concomitant to a higher baseline plasma corticosterone level, and reduced expression of Hsp 70 and 90 in the FC. Pinealectomy reduced the motor activity of the young adult rats during the dark phase and blunted the diurnal activity of old rats. Further, melatonin deficiency related to pinealectomy alleviated anxiety responses to novelty without affecting the expression of Hsp 70 and 90 both in the FC and the hippocampus in young adult and old rats. The impaired activity of the HPA axis was simultaneous to the detected impulsive-like behavior in 18-month rats. The anxiety responses, the circadian rhythms of motor activity, and HPA axis activity were not affected in the middle-aged rats with pinealectomy. Still, this generation exhibited lower levels of the Hsp 70 in the FC.

Our findings for age-related changes in emotional responses in the EPM agree and extended previous reports in different rat strains [1, 2, 6]. However, other studies demonstrated contradictory results [3, 6]. Thus, Sudakov et al. [3] suggested that behavioral alterations related to emotional conditions in rats were evident as early as late adolescence (5-month-old rats). This discrepancy might not be attributed to strain differences since the authors, as mentioned above, also used Wistar rats. In the present study, we used two parameters to measure activity in the EPM (number of entries and total distance traveled), which were indicated as adequate measures of motor activity [1]. In addition, to eliminate the impact of the anxiety factor related to measurement in novelty for this parameter, we also tested motor activity for 24 h in the actimeter, suggesting that at these conditions, rats are well-habituated. The discrepancy between our data and those reported by Torras-Garcia et al. [6] can be explained by our controls being sham-operated rats. In the present study, the age-associated tendency for enhanced anxiety response was also confirmed in two additional tests for anxiety.

It is well-known that there is a close relationship between the plasma profiles of melatonin and the activity of the pineal gland because the indolamine is not stored there [14, 25]. Recently, we reported that removing the pineal gland caused abolished circadian fluctuations of melatonin levels in plasma [26], suggesting that the procedure could be considered an appropriate model of plasma melatonin deficiency. Although many other extrapineal tissues are sources of melatonin, its hormonal activity primarily involves fine-tuning endocrine and other internal “clock” signals synchronized with the external “clock” of the light–dark cycle. The tissues with the highest metabolic activity, like the brain, skin, and gut, are the most critical parts of the body where the indoleamine is also synthesized [27]. Moreover, results from animals and humans reveal that extrapineal melatonin levels can exceed the blood levels of the hormone [28,29,30]. Both glandular (pineal) and cellular sources of detected melatonin in the brain tissue contribute to the higher level in cerebrospinal fluid than other places [31]. Aging is accompanied by a progressive decrease in hormone production with concomitant attenuation of its protective action against oxidative stress and enhanced vulnerability to the onset of neurodegenerative diseases [25].

In the present study, we report the impact of melatonin deficiency, induced in three different generations, on anxiety-like behavior and circadian rhythm of motor activity. The role of endogenous melatonin as a hormone in anxiety behaviors was demonstrated in experimental animals and clinical conditions [32]. We found that young adult and old rats were the most vulnerable to changes in anxiety due to melatonin deficiency, showing less anxiety than the matched sham groups. In contrast, melatonin deficiency did not affect the middle-aged rats. Notably, changed emotional responses in the elderly rats with pinealectomy were not associated with altered motor activity in the novel environment. Therefore, the impaired anxiety behavior of elderly rats with pinealectomy might not be related to impulsivity as an emotional component of motor activity. Pierpaoli and Bulian [15] reported that pinealectomy harmfully affected blood, metabolic, and hormonal parameters in 3-month-old mice but not in 14-month-old rodents. Recently, we confirmed the findings of Pierpaoli and Bulian [15] that melatonin deficiency accelerates aging in young adult rats and can impair several metabolic indices with concomitant elevation of arterial blood pressure, blood glucose, triglyceride, and cholesterol [17, 18]. In the present study, the procedure of pineal gland removal in elderly rats did not confirm the hypothesis of Pierpaoli et al. [15] that pinealectomy in mice older than 14 months no longer affected the “aging program”. Although this suggestion was confirmed for the expression of Hsp 70 and 90 in the FC in the present study, surprisingly, the old rats with melatonin deficiency showed low anxiety expressed with diminished ability to react to dangerous or aversive conditions.

Moreover, unlike Pierpaoli and Bulian [15], who reported that pinealectomy accelerated aging in 3-month-old BALB/c mice with pinealectomy via a shortage of the lifespan, our results showed a lack of impact of blood melatonin deficiency on the lifespan of the youngest rats. In addition, Pierpaoli and Bulian [15] reported that middle age is a crucial period for mice, and pinealectomy caused a delay in lifespan, while melatonin deficiency in 18-month-old mice did not affect this parameter. The discrepancy between the results of Pierpaoli and Bulian [15] in mice and our findings in rats regarding the role of melatonin as a hormone on lifespan needs further clarification and is an open question. Nevertheless, our results indicate that blood melatonin deficiency might not be crucial for lifespan irrespective of the age stage of inducement, suggesting that an extra-pineal source of melatonin is enough to compensate for the lack of hormonal function.

In support of previous reports [33, 34], we demonstrated that plasma corticosterone increased with aging in rats. The discrepancy between these findings and other reports [20, 33,34,35,36] was discussed in the review of Sapolsky [33], suggesting that conflicted results might not be attributed to sex, stress, or time point of detection divergence but to methodology and procedure of blood collecting. The activity of the HPA axis is closely related to responses to stress and might be changed by aging and pathological conditions. In the present study, we report that the HPA axis-associated negative feedback control resulting from stress procedure was intact in the three generations, i.e., young adult, middle-aged, and old sham-operated rats. This finding is consistent with the report of Scaccianoce et al. [34], who demonstrated that corticosterone could suppress the release of ACTH in the pituitary in young and old rats. The adequate adrenocortical reaction after stress in aged rats was also reported by Sapolsky et al. (19,902) [34]. The increased basal corticosterone level in old rats in the present study might be due to impaired feedback control by the hypothalamus [20, 35, 37]. Age-associated melatonin deficiency contributes to high basal corticosterone levels in sham-operated 18-month-old rats compared to the matched 3-month-old rats [38]. We confirmed the result of Oxenkrug et al. [39] and demonstrated that old Wistar rats with pinealectomy had an additional elevation of corticosterone compared to the matched sham-operated rats in basal conditions while young adult and middle-aged groups were not affected. Increased plasma corticosterone from blood melatonin deficiency in elderly rats suggests a disrupted pineal control on adrenal activity.

Further, melatonin deficiency in this group led to hypoactivity of the HPA axis and inadequate response to stress stimulus. The disruption of melatonin production via removing the pineal gland, which is the primary source of this hormone, also changes the mechanism of its synthesis from immune‐competent cells responsible for inflammatory response and involved in the immune-pineal axis [40]. Interestingly, the mechanism of glucocorticoid inhibitory control by the adrenal cortex on melatonin production activated in stress conditions in intact pineal gland should be impaired in pinealectomy and possibly involved in the disturbed HPA axis activity observed in elderly rats. Recently, we reported that young adult rats with pinealectomy showed an impaired circadian rhythm of plasma corticosterone levels [26]. We can speculate that the adaptive response of immune‐competent cells in young adult and middle-aged rats with melatonin deficiency is strong enough to overcome changes in pineal-adrenal gland relationships and maintenance of the HPA axis activity. Therefore, a comparative analysis of the age-related effect of melatonin deficiency on different components of the HPA axis needs further evaluation.

The underlying molecular mechanism in the pinealectomy-induced changes in anxiety response in young adult and old rats was further evaluated by exploring the expression of Hsp 70 and Hsp 90 in homogenates from the FC and the hippocampus. The impaired mechanism of stress-induced expression of Hsp70 was suggested to be a valid biomarker of the aging process [21]. The age-related decline in the expression of Hsp 70 was reported in different rat tissues, including lymphocytes, hepatocytes, hippocampus, and FC [22, 23, 41]. Recently, we demonstrated that while a sub-chronic treatment with melatonin did not affect the expression of the Hsp 70 in basal conditions in the FC and the hippocampus in both the Wistar rats and spontaneously hypertensive rats, the hormone tended to reduce the status epilepticus-induced rise in the shock protein in the hippocampus [42]. Other authors also reported the beneficial ameliorating effect of melatonin treatment on the induction of Hsp 70 in various pathological conditions, including melatonin deficiency induced by pinealectomy [41, 43]. The age-related decline in the expression of the two Hsp (70 and 90) in the FC but not in the hippocampus reported in the present study agrees with the literature data [22, 23] and might be explained by declined hormonal function in plasma.

Conversely, the pinealectomy-induced suppression of Hsp 70 in middle-aged rats might represent an adaptive mechanism against melatonin deficiency in plasma. This suggestion aligns with Pierpaoli and Bulian (2005) hypothesis [15] that pinealectomy in 14-month-old mice might delay the aging process in some aspects. However, our recent studies in rats demonstrated that removing the pineal gland might have a harmful effect, leading to accelerated aging in middle-aged rats, suggesting a more complex role of the hormone across the lifespan. The lack of effect of pinealectomy on the Hsp 70 expression in elderly rats could also be explained by the impaired expression of these proteins in aging.