Vitamin D3 improves spatial memory and modulates cytokine levels in aged rats

Vitamin D3 deficiency is associated with an increased risk of dementia. An association between vitamin D3 deficiency and subjective cognitive complaints in geriatric patients has been previously reported. This study aimed to evaluate the effects of two doses of vitamin D3 on spatial memory (using the Radial Maze) and cytokine levels [tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), interleukin-6 (IL-6), and interleukin-10 (IL-10)] on 2-, 6-, 13-, 22-, and 31-month-old male Wistar rats. Animals were supplemented with vitamin D3 at doses of 42 IU/kg and 420 IU/kg for 21 days. A radial maze test was performed to evaluate spatial memory. After the behavioral test, the frontal cortex and hippocampus were dissected for enzyme immunoassay analyses to measure the cytokine levels (TNFα, IL-1β, IL-6, and IL-10). Our results showed that vitamin D3 supplementation reversed spatial memory impairment at the supplemented doses (42 and 420 IU/kg) in 6-, 13-, and 22-month-old animals and at a dose of 420 IU/kg in 31-month-old animals. The lower dose (42 IU/kg) regulates both pro- and anti-inflammatory cytokines mainly in the frontal cortex. Our results suggest that vitamin D3 has a modulatory action on pro- and anti-inflammatory cytokines, since older animals showed increased cytokine levels compared to 2-month-old animals, and that vitamin D3 may exert an immunomodulatory effect on aging.


Introduction
Human aging is a complex and dynamic process resulting from a combination of environmental, genetic, epigenetic, and stochastic factors. It is characterized by continuous remodeling and low-grade chronic inflammation, a phenomenon termed as inflammaging (Davinelli et al. 2016;Franceschi et al. 2000;Franceschi and Campisi 2014;Minciullo et al. 2016;Prattichizzo et al. 2016). The theory of inflammaging, one of the most recent aspects of aging, concentrates on the immune response and takes into account the activation of chronic low-grade inflammation that occurs with aging (Davinelli et al. 2016;Minciullo et al. 2016;Prattichizzo et al. 2016).
It is a significant risk factor for morbidity and mortality in the elderly and for most age-related diseases that share an inflammatory pathogenesis (Franceschi et al. 2000;Franceschi and Campisi 2014). Several studies have suggested that communication between the brain and immune system is crucial for maintaining central nervous system (CNS) homeostasis. One of the most recognized effects of brain aging is the deregulation of the immune system as a result of the production of reactive oxygen species (ROS) and proinflammatory cytokines (Davinelli et al. 2016;Esiri 2007).
Pro-inflammatory cytokines such as TNF-α and IL-1β can generate an adaptive immune response from astrocytes. Monocyte chemotactic protein-1 (MCP-1 / CCL2) recruits additional immune cells. IL-6 plays a central role in immune responses, hematopoiesis, and acute phase reactions, as well as in neurogenesis and maturation of neuronal and glial cells under normal conditions (Heidary et al. 2014;Hubackova et al. 2012;Simpson et al. 1997).
Epidemiological studies have shown that up to one-third of adults have insufficient levels of vitamin D3, with levels lower than 30 ng/mL in the elderly and lower than 20 ng/mL in the youngest (Briones and Darwish 2012;Nossov et al. 2014). There is a strong association between low vitamin D3 concentration and depression (Collin et al. 2016). In addition, vitamin D3 has been shown to be involved in processes associated with neurogenesis and have well-known roles in calcium metabolism, bone health, and the immune system (Dougherty et al. 2016;Pang et al. 2016). An association has been demonstrated between low concentrations of vitamin D3 and impairment of cognitive functions such as executive function, memory, and orientation, as well as with the incidence of dementia and Alzheimer's disease (Annweiler et al. 2009;Llewellyn et al. 2009).
Vitamin D3 plays an immunomodulatory role in the CNS (Casaccia-Bonnefil et al. 2008;Al-Harbi et al. 2017). Clinical studies have demonstrated an association between low vitamin D3 level and episodic memory disorders. It has also been shown that higher vitamin D3 levels are associated with better cognitive performance. Vitamin D3 deficiency is also associated with an increased risk of dementia and subjective cognitive complaints in geriatric patients (Annweiler et al. 2013;Littlejohns et al. 2014;Llewellyn et al. 2009;Soni et al. 2012;Tot Babberich Ede et al. 2015).
We hypothesized that vitamin D3 protects against memory impairment and neuroinflammation induced by aging. The present study aimed to evaluate the effects of two doses of vitamin D3 on spatial memory and cytokine levels in adult and old Wistar rats.

Drugs
The drug used was: Vitamin D3. It was obtained from Farmasa S/A, Pharmacotherapy American Laboratory, São Paulo, Brazil (Adera D3®-colicalciferol-3,300 IU/ml). It was diluted in water every three days at doses of 42 or 420 IU/kg at a volume of 1 mL/kg body weight. The doses of vitamin D3 and the time points were chosen based on a previous study (Briones and Darwish 2012). The vehicletreated control group was simultaneously administered water at a volume of 1 mL/kg body weight.

Experimental design
The animals were divided into three experimental groups; two of the groups were administered vitamin D3 at doses of 42 or 420 IU/kg for 21 days by gavage, and the control group was administered water. The gavages were scheduled daily at 10:00 a.m. The animals underwent behavioral tests from the 18 th to the 22 nd day. At the end of the behavioral test on the last day (22 nd day), the animals were euthanized. The frontal cortex and hippocampus were dissected for biochemical analyses. The rats were randomly divided into three groups (initially 13 rats in each group) for every age group (2, 6, 13, 22, and 31 months) (Fig. 1).
The study included 15 experimental groups. Considering that the aging process increases mortality, we had variation in the n of the experimental groups because the older the age, the higher the mortality. The 31-months-old animals had a mortality rate of approximately 69%. A study Fig. 1 Experimental design. Treatment protocol in all ages (2, 6, 13, 22, and 31 months) conducted by Phillips et al. (2010) observed that 31-monthold rats had about 60% mortality which is consistent with our study.

Radial arm-maze task
The radial maze test was performed to evaluate spatial memory, on the 18 th day of treatment. The radial maze apparatus had 8-arms, which were numbered from 1 to 8 (48 × 12 cm) and extended radially from the central area (32 cm diameter). It was placed 50 cm above the floor, and geometric shapes were positioned in the straight arms where the food was (visual cues). On the zero day, each animal was placed in the apparatus for a total of 10 min, allowed only to explore, and then returned to its cage and kept on a restricted diet. Their body weight was maintained at 85% of their free-feeding weight over a period of one week, with water available ad libitum. On the 1 st day the animals were placed in the apparatus, where food (chocolate cereal) had already been deposited in four of the eight arms. The foodbearing arms had visual cues at the end of each arm. Over a period of 10 min, entry into each arm (total errors to find food) and the time each animal took to find the four pieces of cereal were recorded (latency to find food). The same test was conducted over four consecutive days, with one trial per day (Foyet et al. 2011;Hritcu et al. 2012). Behaviors were analyzed by a trained researcher blinded to the treatments. After the last test and 24 h after the last administration of vitamin D3, animals were euthanized.

Euthanasia and tissue preparation
The animals were euthanized on the 22 nd day of treatment by decapitation. They waited for euthanasia in an adjacent room to prevent them from smelling the blood, and the guillotine was washed in running water between each euthanasia.
The brains were rapidly removed, placed on an ice-cold cutting board, and washed with saline. After removal of the meninges, hippocampi and frontal cortex were extracted, snap-frozen in liquid nitrogen, and stored at -80 °C until biochemical analysis. Both hippocampi were removed. The frontal cortex was dissected according to the method of Paxinos and Watson (1998)

Enzyme-linked immunosorbent assay
Frontal cortex and hippocampal samples were solubilized in PBS (pH 7,4). Cytokines (TNF-α, IL-1β, IL-6, and IL-10) were quantified using enzyme immunoassay kits (R&D Systems, Minneapolis, MN, USA) as recommended by the manufacturer (described below). Microtiter plates (96 flat bottom wells) were incubated overnight with the capture antibody. Subsequently, the plates were washed three times with wash phosphate buffer and then blocked with 1% phosphate buffer solution for 1 h. Afterwards, the homogenized samples and standards, which were both diluted in buffer solution, were incubated for 2 h. Subsequently, the plates were washed thrice with wash buffer and then incubated with the detection antibody for another 2 h. After the washes, streptavidinconjugated peroxidase was incubated for 20 min, followed by the substrate (hydrogen peroxide and tetramethylbenzidine, 1:1). The reaction was stopped by 2 N sulfuric acid. The absorbance of the plates was read at 450 nm using a spectrophotometer. Total protein was measured according to the method described by Lowry et al. (1951), using bovine serum albumin as a standard.

Statistical analysis
Statistical analyses were performed using Statistica software 8.0 (StatSoft Inc., Tulsa, USA). Data from the ELISA analyses were evaluated by one-way analysis of variance (ANOVA), followed by Duncan's test when p values were < 0.05. Mauchly's test of sphericity was used (assumption of violated sphericity), and repeated measurements using the one-way ANOVA test were performed to assess radial arm-maze task data following Duncan's post hoc test. The data were reported as mean ± SEM, and p values < 0.05 were considered statistically significant.
The results of the radial arm maze test are shown in Figs. 2A and B. Figure 2A shows the results of the radial maze test as the time taken to find food, and Fig. 2B presents the results for the number of errors in finding food. There was a significant difference in the ANOVA test when the latency time to find the reward [F (3,42) = 2,19; p < 0.001)] and the total errors to find the reward were observed [F (3,42) = 27,87; p < 0.05].
Duncan's post-hoc test revealed differences in the time ( Fig. 2A) to find the reward in the control groups, as well as in 2-months-old animals with doses of 42 and 420 IU/ kg of vitamin D3. They were significant in the 2 nd , 3 rd , and 4 th days; thus, the animals had a conserved spatial memory, as each day, they reduced the time to find the reward. However, the animals at 6 and 13 months of age presented spatial memory impairment in the control group which was prevented by doses of 42 and 420 IU/kg of vitamin D3, but only on the 4 th day of the test. Animals at 22 and 31 months of age also presented spatial memory impairment in both the control group, and the group treated with 42 IU/kg of vitamin D3. However, the dose of 420 IU/kg was able to prevent this, when tested on the 3 rd and 4 th day ( Fig. 2A).
Evaluating the total errors (Fig. 2B), Duncan's post hoc test showed differences at the age of 2 months on the 2 nd , Effect of vitamin D3 treatment (42 and 420 IU/kg) for 21 days in animals 2 (6 to 9 animals per group), 6 (7 to 8 animals per group), 13 (4 to 9 animals per group), 22 (5 to 6 animals per group), and 31 (4 to 5 animals per group) months old in spatial memory evaluated by the radial maze task. (A) Time to find the food and (B) total errors to find the food on 1 st , 2 nd , 3 rd and 4 th test days; n = 4-9 animals per experimental group. Data are expressed as mean ± SEM. Data that show * p < 0.05 are considered significant when compared to the 1 st day of test within the experimental group 3 rd , and 4 th days when compared to the 1 st day in the water group. The group treated with 42 IU/kg of vitamin D3 had a better performance on the 4 th day of the test, and the group treated with a dose of 420 IU/kg obtained better performance on the 3 rd and 4 th days of the test. The animals at 6, 13, 22, and 31 months that received water presented memory impairment, and the number of errors on the test days was not reduced. At the age of 6 months, there was a reduction in the number of errors on the 4 th day in the groups treated with 42 and 420 IU/kg of vitamin D3. In contrast, the 13-monthsold animals treated with doses of 42 and 420 IU/kg of vitamin D3 showed decreased errors on the 3 rd and 4 th test days when compared to the 1 st day within each treated group. At the age of 22 months, the 42 IU/kg vitamin D3 dose reduced the number of errors on the 3 rd and 4 th days of the test, but the group treated with 420 IU/kg only decreased the errors on the last day. However, animals aged 31 months showed cognitive impairment in all treatments, and there was no decrease in errors within any group (Fig. 2B).
The IL-10 levels in the frontal cortex [F (1,14) = 6.63; p < 0.01] and hippocampus [F (1,14) = 37.47; p < 0.01] are shown in Fig. 6A and B. At the ages of 2 [42 IU/ kg (p = 0.31); 420 IU/kg (p = 0.07]) and 6 [42 IU/kg (p = 0.36); 420 IU/kg (p = 0.43]) months no differences were seen in the frontal cortex when compared to the control (Fig. 6A). At the age of 13 [42 IU/kg Fig. 5 Effect of vitamin D3 treatment (42 and 420 IU/kg) for 21 days on IL-6 levels in brain structures of the frontal cortex and hippocampus of Wistars rats. Figure 5A shows levels of IL-6 (pg/mL) in the frontal cortex; 5B shows levels of IL-6 (pg/ml) in the hippocampus. Data are expressed as mean ± SEM of 3-6 animals per group, # p < 0.05, ## p < 0.01 when compared to the water group within the same age and * p < 0.05, ** p < 0.01 when compared to the water group of 2 months of age (p < 0.001); 420 IU/kg (p > 0.001]) and 31 [42 IU/kg (p < 0.05); 420 IU/kg (p > 0.01]) months, the control group showed an increase in IL-10 levels, and the two doses of vitamin D3 (42 and 420 IU/kg) decreased these levels in both groups (Fig. 6A). At 22 months, the dose of 42 IU/kg (p < 0.05) decreased IL-10 levels compared to the control; however, the same did not occur at 420 IU/kg (p = 0.85) (Fig. 6A). In the hippocampus (Fig. 6B), animals at 2 months of age showed increased IL-10 levels at both doses of 42 IU/kg (p < 0.05) and 420 IU/kg (p < 0.01) when compared with the control. At 22 months (Fig. 6B), contrary to what was observed in the frontal cortex (Fig. 6A), the dose of 42 IU/kg (p < 0.05) increased IL-10 levels when compared to the control, and the dose of 420 IU/kg (p = 0.23) had no significant effect. In the hippocampus, the ages of 6, 13, and 31 months at both doses [42 IU/kg (p = 0.72; p = 0.24; p = 0.13) and 420 IU/kg (p = 0.66; p = 0.19; p = 0.12]), respectively, did not present differences when compared to the control within each age group (Fig. 6B). Table 1 summarizes the effect of treatment with vitamin D3 (42 and 420 IU/kg) for 21 days in animals at 2, 6, 13, 22, and 31 months of age on spatial memory and the levels of TNF-α, IL-1β, IL-6, and IL-10 in the cerebral structures of the frontal cortex and hippocampus. Fig. 6 Effect of vitamin D3 treatment (42 and 420 IU/kg) for 21 days on IL-10 levels in brain structures of the frontal cortex and hippocampus of Wistars rats. Figure 6A shows levels of IL-10 (pg/mL) in the frontal cortex; 6B shows levels of IL-10 (pg/ml) in the hippocampus. Data are expressed as mean ± SEM of 3-6 animals per group. # p < 0.05, ## p < 0.01 when compared to the water group within same age and * p < 0.05, ** p < 0.01 when compared to the water group of 2 months of age

Discussion
The effects of vitamin D3 on spatial memory were evaluated using the radial maze task (Dellu et al. 1997;Lee et al. 2014). In rodents, it has been reported that memory and spatial functions decline during natural aging (Mora-Gallegos et al. 2015). It is important to emphasize that the present study is the first to evaluate vitamin D3 treatment in animals at 13, 22, and 31 months old (natural aging). The animals at 2 months of age did not present spatial memory impairment, as shown by the reduced time and errors in finding food. The animals at 6, 13, 22, and 31 months of age did not learn the task, probably because of spatial memory impairment caused by aging. However, when treated with vitamin D3 (42 and 420 IU/kg), animals at 6, 13, and 22 months of age learned the task. In addition, in rats at 31 months of age, only the 420 IU/kg dose was able to improve performance by reducing the time to find food on the 3 rd and 4 th day of the test. The results demonstrated that vitamin D3 improved spatial memory in these animals.
A study using transgenic F344 male rats, a lineage of accelerated aging mice, showed that vitamin D3 (42 IU/ kg intraperitoneally) for 21 days improved spatial memory in a water maze task in 20-months-old rats (Briones and Darwish 2012). In another study conducted in 2014, F344 male rats aged 11 to 13 months received diets containing low (100 IU), medium (1,000 IU), or high (10,000 IU) levels of vitamin D3 for 5 to 6 months. The results showed that the group supplemented with high levels of vitamin D3 performed the water maze task extremely well and reached the goal in half the time and distance taken by the other two groups (Latimer et al. 2014). In addition, another study was performed in which the 10-day Senescence Accelerated Mouse-Prone 8 (SAMP8) received 500 IU/kg of vitamin D3 alone or in combination with resveratrol (43 mg/kg) in their diet for 24 weeks. Supplementation with vitamin D3 and resveratrol reversed cognitive impairment in the water maze task in 34-week old SAMP8 mice, and the combined vitamin D3 supplementation with resveratrol was more effective than isolated resveratrol intervention (Cheng et al. 2017). These data are consistent with the results of the present study.
One of the most recognized effects of brain aging is deregulation of the immune system as a result of uncontrolled production of reactive oxygen species and proinflammatory cytokines (Davinelli et al. 2016;Esiri 2007). With advancing age, the body develops a process known as "chronic inflammation" or inflammaging (Gruver et al. 2007;Watson et al. 2017).
To investigate the immunoprotective effects of vitamin D3, we evaluated the TNF-α, IL-1β, IL-6, and IL-10 levels. Vitamin D3 was able to reduce cytokine levels at 13 months of age, since TNF-α, IL-1β, and IL-10 levels decreased in the frontal cortex by both doses of vitamin D3, and TNFα Table 1 The results of the effect of vitamin D3 treatment (42 and 420 IU/kg) for 21 days in animals 2, 6, 13, 22, and 31 months old in the radial maze test (spatial memory), and the levels of TNF-α, IL-1β, IL-6, and IL-10 in cerebral structures of the frontal cortex and hippocampus of Wistar rats. To analyze the results of spacial memory we used three symbols 1. * used when the control (water) has no memory impairment; 2. ҉ represents vitamin D3 doses of 42 and 420 IU/kg with p < 0.05 when compared to the water group within the same age group and indicated no memory impairment; 3. × , memory impairment; ↑, increased levels of cytokines; ↓, decreased levels of cytokines; -, no changes in cytokine levels; RM = Radial Maze; FC = Frontal Cortex; HIP, hippocampus levels decreased in the hippocampus only at a dose of 42 IU/ kg. The inflammatory cytokines TNFα, IL-1β, and IL-10 increased in the frontal cortex and hippocampus due to aging at 13 months of age. The immunomodulatory effects of vitamin D3 were observed principally in the frontal cortex. At 22 months of age, vitamin D3 at a dose of 42 IU/kg decreased pro-inflammatory cytokine IL-6 and anti-inflammatory IL-10 levels in the frontal cortex, and the same dose was able to increase the levels of IL-6 and IL-10 in the hippocampus compared to the control. The 31-months-old animals that received water showed increased levels of IL-1β and IL-10, which were reversed by both doses of vitamin D3. Additionally, vitamin D3 42 IU/kg reduced IL-6 levels in the frontal cortex and only a dose of 420 IU/kg reduced IL-6 levels in the hippocampus. However, 42 IU/kg of vitamin D3 reduced IL-1β levels in the hippocampus, and a dose of 420 IU/kg increased the levels of this in the hippocampus.
In the present study, animals of advanced age demonstrated an increase in cytokine levels, and in most cases, vitamin D3 treatment was able to modulate inflammatory cytokine levels and prevent memory impairment. In a study using F344 rats aged 6 and 20 months, vitamin D3 injection at a dose of 42 IU/kg reduced the age-related pro-inflammatory state and amyloid burden, suggests these vitamin D3 effects are related to the memory improvements observed in elderly rats (Briones and Darwish 2012). This study corroborates data from the present study, since vitamin D3 has a modulatory effect on cytokines.
However, in the present study, we showed unexpected results since vitamin D3 treatment increased the levels of some pro-inflammatory cytokines. A dose of 42 IU/kg increased IL-6 levels in the hippocampus in 22-months-old rats. In addition, the dose of 420 IU/kg increased IL-1β in the frontal cortex and hippocampus of 2-months-old rats, IL-1β in the hippocampus of 31-months-old rats, and TNF-α in the frontal cortex of 22-months-old rats. Vilela et al. (2017) observed an increase in IL-1β and TNF-α in the hippocampus of 24-months-old rats subjected to strength training, although spatial memory was improved. The authors suggested that the increase in these inflammatory cytokines may be related to memory improvement (Vilela et al. 2017) since they play an active role in neuronal development and neuroplasticity (Khairova et al. 2009;Tonelli and Postolache 2005).
IL-1β has been shown to play a putative role in hippocampal-dependent memory formation, in addition to its inflammatory effects. It can play a dual role in this memory process (Goshen et al. 2007;del Rey et al. 2013;Barter et al. 2021). Moreover, IL-6 may also be critical for memory processing. A study performed by Li et al. (1997) showed that IL-6 modulates synaptic potentiation through IL-6 receptormediated processes in the hippocampus, probably by affecting post-and presynaptic sites in the CA1 region. Another study conducted by del Rey et al. (2013) showed that IL-1β and IL-6 are overexpressed during long-term potentiation (LTP) maintenance and are causally related to an increase in synaptic strength. These cytokines are overexpressed in defined regions of the hippocampus during learning (del Rey et al. 2013). These results corroborate our results that showed vitamin D3 induced an increase in IL-1β (at a dose of 420 IU/kg) and IL-6 (at a dose of 42 IU/kg) levels in the hippocampus, indicating a protective effect of vitamin D3.
Furthermore, considering that certain basal levels of TNF-α are required for the development of normal cognition, this cytokine can also be related to the maintenance of synaptic scaling and thus influence learning and memory formation (Camara et al. 2013) corroborating our results that showed that vitamin D (at a dose of 420 IU/kg) increased TNF-α levels in the frontal cortex of 22-months-old rats.
Our results showed that throughout the aging process (mainly in rats aged 22 and 31 months), a different response was observed with varying doses of vitamin D3 (42 or 420 IU/kg) in different regions of the brain (hippocampus or cortex) with reference to levels of different cytokines (IL-1β, IL-6 or TNF-α). Therefore, the current study's findings showed that vitamin D3 supplementation was associated with the attenuation of several age-related inflammatory changes in the brain, suggesting a protective influence on the aging brain. In addition to its potential direct neuroprotective effects, vitamin D3 modulates inflammation and counteracts age-related low-grade brain inflammation.
In our opinion, future studies are needed to compare male and female rats to improve the accuracy and reliability of these results. In addition, since older rats have higher mortality, it would be interesting to use more animals in the aged groups in future studies.
Our results suggest that vitamin D3 modulates pro-and anti-inflammatory cytokines and prevents spatial memory impairments, and support the hypothesis that vitamin D3 exerts an immunomodulatory effect on aging. It is necessary to conduct more studies, since our results showed that the regulation of cytokines in the brain may be a complex process.