High-protein and low-calorie diets improved the anti-aging Klotho protein in the rats’ brain: the toxic role of high-fat diet

Background In the current study, our specific aim was to characterize the Klotho protein and expression levels in the hippocampus and prefrontal cortex of old rats treated with different diets (high-fat, high-protein, low-calorie, high-protein and low-calorie). Methods Rats were treated with high-fat, high-protein, low-calorie, low-calorie high-protein diets for 10 weeks and then behavioral and molecular assessments were evaluated. Results Statistical analysis showed the percentage of open arm time was increased in the high-protein, low-calorie and low-calorie high-protein groups compared with old control (old-C) rats. The percentage of open arm entries was increased in the low-calorie and low-calorie high-protein group compared with old-C rats. The body weight and serum triglyceride were decreased in the low-calorie and low-calorie high-protein groups in comparison to control old rats. Low-calorie and low-calorie high-protein treatments statistically enhanced caspase-3 level compared with old-C rats in the hippocampus and prefrontal cortex. Treatment of old rats with high-protein, low-calorie and low-calorie high-protein could increase Klotho-α level compared with control old rats. The levels of Klotho-α, c-fos and brain-derived neurotrophic factors were decreased in the low-calorie high-protein group in Klotho inhibitor's presence compared with the low-calorie high-protein group. Conclusion According to our findings, Klotho-α level was reduced in old rats. Low-calorie, high-protein and particularly low-calorie high-protein diets increased this protein level and consequently increased neuronal plasticity and improved memory function. Graphic abstract


Experimental design
This study is designed into two experimental designs. In the first experimental design, 20-month-old and 6-month-old male rats were randomly put in six groups. All rats in each group were housed individually (n = 7): (1) adult control (adult-C), adult rats were treated with control diet for 10 weeks, (2) old control (old-C), old rats were treated with control diet for 10 weeks, (3) high-fat diet, old rats were treated with high fat food for 10 weeks, (4) high-protein diet, old rats were treated with high-protein food for 10 weeks, (5) low-calorie diet, old rats were treated with low amount of control diet for 10 weeks, (6) low-calorie high-protein diet, old rats were treated with low amount of high-protein food for 10 weeks. In the second experimental design, 20-month-old male rats were randomly put in four groups. (1) old control (old-C), old rats were treated with control diet for 10 weeks, (2) low-calorie high-protein diet, old rats were treated with low amount of high-protein food for 10 weeks, (3) old-C and d-saccharic acid 1,4-lacton diet, old rats were treated with low amount of high-protein food for 10 weeks and received d-saccharic acid 1,4-lactone twice per week for 10 weeks, (4) low-calorie diet and d-saccharic acid 1,4-lacton diet, old rats were treated with low amount of high-protein food for 10 weeks and received d-saccharic acid 1,4-lactone twice per week for 10 weeks. The first and second experimental design groups are the same second and sixth groups in the first experimental design.

Dietary protocol
5 g/100BW of control diet was given to adult and old control rats each day for 10 weeks. In a high-protein diet, rats were given 5 g/100BW of high-protein food each day for 10 weeks [11]. In a high-fat diet, rats were given 5 g/100BW (210 kcal (kg d)) of high fat food each day for 10 weeks [12]. In a low-calorie diet, rats were given 3.5 g/100BW (140 kcal (kg d)) of control diet each day for 10 weeks. A decreased (30%) calorie diet was applied with no uneven reduction of any dietary component in lowcalorie diet [13]. In a low-calorie high-protein diet, rats were given 3.7 g/100BW (140 kcal (kg d)) of high-protein food each day for 10 weeks. Table 1 shows each diet's ingredients. The formulation of the diet was done based on the nutrient necessities of adult rats according to the guidelines of the American Institute of Nutrition (AIN-93 M). Behparvar Company (Tehran, Iran) produced normal, high-protein and high-fat diets. Table 1 shows each diet composition. The rats were given d-saccharic acid 1,4-lactone, a Klotho inhibitor orally two times per week for 10 weeks (80 mg/kg Body weight) [14].

Behavioral tests Elevated plus maze (EPM) test
The wooden apparatus having two open and two closed arms (each 50 × 10 cm) with 40 cm high sidewalls, was placed 50 cm above the floor [15]. Following 10 weeks of treatment, rat was put in the center of raised plus maze with its head facing an open arm and left undisturbed for 5 min. All behavioral tests were done from 10:00 a.m. to 14 [16,17].

Open field test
The open field opaque acrylic box (50 × 50 × 40 cm) was put in an isolated room without any objects or clues. The box floor was separated into 10 numbered squares, roughly 16 × 16 cm 2 each for central or parietal localization. Following 10 weeks of treatment, the locomotion of each rat in the box for 5 min was tested and filmed by a camera fixed on the ceiling of the room (Maze router, Tabriz, Iran). Both maze router analysis did the scoring, and manually that was expressed as "Moving distance (meters)" and "Time spent in the central square (s)" [16,17].

Novel object recognition test (NOR)
NOR was done in a square open field (50 × 50 × 40 cm). An investigator blinded to the treatment groups performed the test within 3 days. The habituation of the rats to the empty arena for 20 min was done 68 days after treatment. All behavioral tests were done from 10:00 a.m. to 14:00 p.m. On day 69, rats were let to go back to the arena with two identical objects for 6 min. On day 70, one of two familiar objects was substituted with a new one with different shapes, colors, and materials. Rats wandered in the arena for 6 min, and total time spent for each object was logged. Exploring an object was set as nosing and sniffing at it at a distance less than 2 cm. The objects and field were cleaned with 75% ethanol between each trial. On the last day, the used object was substituted, and the rats were monitored to find the possible position preference. Data was shown as the percentage of recognition index (RI) determined as follows: [time spent in exploring the novel object /time spent in exploring the two objects] × 100 [18].

Body weight, blood glucose and triglyceride level
A digital balance (Farooq chemical, Tehran, Iran) was used for assessing body weight on days 1 and 70. The blood samples were gathered via the tail of animals. On day 1 and 10 weeks following treatment, the blood glucose monitoring kit (Glucocard01, Japan) was applied for detecting the blood glucose level. Afterward, the blood samples underwent centrifuging at 2500 rpm for 5 min for gathering the supernatants for analysis. The triglyceride detection kit (Mancompany, Tehran, Iran) was applied for assessing triglyceride levels.

Western blot analysis
The hippocampi and prefrontal cortex were dissected and flash-frozen in liquid nitrogen and stored at − 80 °C. Before Western blot analysis, all samples' homogenization was done in the suitable lysis buffer, [19] and total protein extract was prepared by centrifuging in 15,000 rpm for 5 min. Bradford's method [20] was applied for measuring the protein concentration in the supernatants. After loading standardized lysates equivalent to 60 µg of protein on SDS-12.5% poly acrylamide gel electrophoresis and transferring to PVDF membrane (Chemicon Millipore Co. Temecula, USA), blots were blocked in 2% Electrochemiluminescence (ECL) advanced kit blocking reagent (Amersham Bioscience Co. Piscataway, USA) and probed with Klotho-α, FGF23, BDNF and c-fos (1/1000, MyBioSource Inc. CA, USA) antibodies overnight. Then, the membranes' incubation was done with rabbit IgG-horseradish peroxidase-conjugated secondary antibody (1/3000, Cell Signaling Technology Co. New York, USA) that was directly noticeable by chemiluminescence kit reagent (Amersham Bioscience Co. Piscataway, USA). For detecting β-actin as an internal control, stripping buffer (pH 6.7) was used to strip the blots and then probing was done with anti β-actin antibody (1/1000, Cell Signaling Technology Co. New York, USA) [19].

Caspase-3 and TGF-β1 level
Caspase-3 and TGF-β1 Elisa assay kits (Abcam, Cambridge, MA, USA) were applied based on the manufacturer's protocol to determine caspase-3 level. A specific lysis buffer was used to lyse the hippocampi and prefrontal cortex tissues, and lysate underwent centrifuging at 15,000 rpm for 5 min at 4 °C, and the Bradford dye method [21] was applied for determining the protein concentration in the supernatants. Caspase-3 and TGF-β1 activity was assessed by measuring the absorbance at 405 nm.

Real time PCR
Step One Plus Real-Time PCR System (Applied Biosystems) was applied for real time PCR reaction for hippocampi and prefrontal cortex of Klotho α gene. Mixing of 2 μl of cDNA, 2 μl of primer and SYBR Green Master Mix (Takara, Japan) was done following manufacture protocol in total volume of 20 μl. The annealing temperature for all primer pairs was set at 58 °C. The PCR product specificity was verified by confirming a single peak in the melting curve and visualizing 2% agarose gel with ethidium bromide in gel documentation. The cycle at which the sample fluorescence faced a predetermined threshold (Ct) considerably beyond the background was used for determining the quantity of cfos and bdnf gene for each sample. Samples were tested in duplicate and the mean was used for further analysis. The housekeeping gene (β-actin) was applied for normalizing all the data of sample and control groups.

Sections preparation and immunofluorescence
PBS was perfused through the ascending aorta with PBS, followed by fresh 4% paraformaldehyde in PBS (pH 7.4) (n = 4). The brains were removed quickly and put in 4% paraformaldehyde in PBS for 24 h. After embedding the tissues with paraffin, they were sectioned at a thickness of 5 μm. Then, incubation was done with normal goat serum (1:100, Cell Signaling Technology Co. New York, USA); next, Klotho-α-specific antibody (1:100 dilution) covered the sections overnight at 4 °C followed by fluorescent secondary antibody Alexa Fluor 488-conjugated anti-rabbit IgG (1:500, Molecular Probes, Invitrogen, CA). A fluorescence microscope (Olympus IX71, Japan) was used for capturing the figures.

Statistical analysis
Obtained data were expressed as mean ± S.E.M for behavioral and mean ± S.D for molecular data. Immunofluorescence staining and Western blotting blots were analyzed by quantification of densitometry data using ImageJ software (National Institutes of Health, Bethesda, MD, USA). Kolmogorov-Smirnov test was performed to find normal assumption of data. Behavioral and Western blot results were analyzed statistically by One-way ANOVA followed by Tukey's multiple comparison; body weight, blood glucose and serum triglyceride were analyzed by repeated measure test. Correlations of assessed proteins were measured by Pearson coefficients correlation and Principal component analysis. All mentioned parameters were analyzed by SPSS software (Version 21). Data of gene expression experiments were analyzed by Relative Expression Software Tool (REST)-XL (Version 2) [22], which is used to determine significant differences in relative expression levels between sample and control groups. P value less than 0.05 is considered statistically significant.

Results
High-fat diet has negative and low-calorie high-protein diet has positive effect on anxiety-like behavior, locomotion, and working memory in the old rats . Percentages of OAT, OAE, and time of center occupancy were decreased in old-C rats compared with adult-C (P < 0.001). Percentage of OAT was increased in the high-protein, low-calorie and low-calorie high-protein groups compared with old-C rats. Percentage of OAE was increased in the low-calorie (P = 0.004) and low-calorie high-protein (P < 0.001) group compared with old-C rats. Percentage of OAE was increased in the low-calorie high-protein groups compared with high-protein group (P < 0.05). Time of center occupancy was increased in the low-calorie high-protein group compared with control old rats (Fig. 1d, P = 0.005) and high-protein group (P < 0.05). Figure 1e, f show that the old rats' recognition index was decreased compared to adult rats (n = 7, P < 0.001). Statistical analysis showed that high-fat diet decreased the recognition index in the old rats compared with control diet group. Treatment of old rats with lowcalorie high-protein diet could enhance the recognition index compared with control diet group (P = 0.003). Table 2 shows the body weight, blood glucose, and serum triglyceride level in the experimental groups before and after 10 weeks of treatment (n = 7). Before treatment, body weight, blood glucose and serum triglyceride of old-C rats were lower than adult-C rats (P < 0.05). Highprotein treatment increased body weight (P = 0.005) and decreased serum triglyceride (P = 0.002) in comparison to control old rats. Data analysis showed the body weight and serum triglyceride were decreased in the low-calorie and low-calorie high-protein groups in comparison to control old rats. The blood glucose was decreased in the low-calorie and low-calorie high-protein compared with control old rats (P = 0.03).

Different diets change caspase-3 and TGF-β1 level in the hippocampus and prefrontal cortex
Caspase-3 and TGF-β1 levels were detected in the hippocampus and prefrontal cortex (n = 4, Table 3). Data reported that caspase-3 level was increased in the hippocampus of old-C rats compared with adult-c rats (P < 0.001). In addition, the TGF-β1 level was reduced in the hippocampus and prefrontal cortex of old-C rats compared with adult-c rats (P < 0.001). High-fat treatment increased caspase-3 and TGF-β1 level compared with old-C rats in the prefrontal cortex. Low-calorie (P = 0.02) and low-calorie high-protein (P < 0.001) treatments reduced caspase-3 level compared with old-C rats in the hippocampus and prefrontal cortex. High-protein (P = 0.02), low-calorie (P = 0.04) and lowcalorie high-protein (P < 0.001) treatments improved TGF-β1 level compared with old-C rats in the hippocampus and prefrontal cortex. High-protein low-calorie Klotho-α and FGF23 were improved by high-protein, low-calorie and low-calorie high-protein diets in the hippocampus and prefrontal cortex of old rats As shown in Fig. 2a, b, the amount of Klotho-α was reduced in the old-C rats compared with adult-C group (P < 0.001). Mean of Florence intensity in the prefrontal cortex showed high-fat diet decreased Klotho-α in comparison to control old rats (P = 0.02). Treatment of old rats with high-protein, low-calorie and low-calorie high-protein could increase Klotho-α level compared with control old rats (P < 0.001). Figure 3c represented a blot of Klotho-α and FGF23; densitometry analysis in Fig. 2d, e reported high-protein, low-calorie and low-calorie high-protein increased Klotho-α and FGF23 levels compared with control old rats in the both hippocampus and prefrontal cortex. Also, Klotho-α level was increased in low-calorie high-protein group compared with highprotein group in the prefrontal cortex (P < 0.05).

High-protein, low-calorie and low-calorie high-protein diets could improve c-fos and BDNF level in the brain of old rats
RT-PCR data was used to evaluate c-fos and BDNF gene expression in the hippocampus and prefrontal cortex of experimental groups (Fig. 3a, b). Results showed that both c-fos and BDNF expression were decreased in the old-C rats compared with adult-C rats (P < 0.001). Treatment of old rats with high-protein, low-calorie, and lowcalorie high-protein diets improved c-fos level compared with control old rats in the hippocampus and prefrontal cortex. c-fos level was increased in low-calorie high-protein group compared with high-protein group in the both hippocampus (P < 0.01) and prefrontal cortex (P < 0.05). Level of BDNF was increased in the high-protein and low-calorie high-protein groups compared with control old rats in the hippocampus and prefrontal cortex. The high-fat diet reduced hippocampal BDNF in comparison to control old rats (P < 0.001). Low-calorie diet increased prefrontal cortex BDNF compared with control old rats (P < 0.001). BDNF level was increased in low-calorie high-protein group compared with high-protein group in the prefrontal cortex (P < 0.001).  Table 4 shows that Pearson coefficients correlation showed a correlation between Klotho protein level in the brain and all assessed proteins or genes (FGF23, TGFβ, caspase-3, c-fos and BDNF). Data analysis indicated there is significant correlation between Klotho and all mentioned factors.

Discussion
It has been well accepted that consumption of low-quality diets are associated with higher frailty in humans and animals [23]; moreover, it is believed that some dietary patterns such as high-protein and calorie-restricted diets can suppress neurodegeneration with aging [24,25]. However, little is known about the influence of dietary patterns on the molecular process of aging. Herein, we showed the differences between adult and old rats in terms of movement and anxiety disorders, and focused on three types of diets (high-fat, low-calorie, high-protein and low-calorie high-protein). We found deleterious effects of high-fat diet and beneficial influence of low-calorie, high-protein and low-calorie high-protein diets on the aging process. Low-calorie and low-calorie high-protein diets could reduce body weight, blood glucose and serum triglyceride level. On the other hand, high-protein treatment only decreases serum triglyceride levels. Arbo et al. [26], confirmed our data and demonstrated old rats have metabolic syndrome, and a high-calorie diet . Rats treated with high-fat, high-protein, low-calorie, low-calorie high-protein diets for 10 weeks. Control adult and old rats treated with rodent standard pellet. After behavioral tests, rats were euthanized and brains of three rats in each groups were collected for RT-PCR (n = 4 and technical repeat for each n is 3). Data were presented as Mean ± S.D. ***P < 0.001 ver. Adult-C, # P < 0.05, ## P < 0.01, ### P < 0.001 ver. Old-C. $ P < 0.05, $$ P < 0.01, $$$ P < 0.001 ver. High-protein. Adult-C adult control rats, Old-C old control rats increases the serum triglyceride and body weight, showing the importance of a low-calorie diet against aging. Besides, studies proved that calorie-restriction decreases and high-fat diet increases serum triglyceride and blood glucose [27,28]. A randomized trial showed a high-protein diet could improve body weight, blood glucose and serum triglyceride [29]; but our observation revealed a high-protein diet only could decrease serum triglyceride. Many studies indicated different protein types influence energy intake and body weight in rats receiving a highprotein diet [30,31], and high-protein diet with either moderate or low carbohydrate has a favorable impact on metabolic health [32], which is in parallel with our finding that low-calorie high-protein diet has more effect rather than high-protein diet in metabolic factors. Also, it has been reported high-protein diet impaired liver function [33]; however, in our study, not only all three factors-high-protein, low calorie and low-calorie highprotein diets-have no degenerative role on liver function, but also, klotho-α level increased in liver, which showed the protective role of these three diets on body metabolism (under publishing data).
Results of the present study confirmed old rats have less anxiolytic, locomotive and memory function in and BDNF (f) bands were measured in the hippocampus and prefrontal cortex. Data were presented as Mean ± S.D. ***P < 0.001 ver. low-calorie high-protein group. Adult-C adult control rats, Old-C old control rats comparison to young rats, which highlighting the impairment of cognitive function in old rats, [34] and high-fat diet worsens anxiety-like behavior and working memory in old rats, but, in contrast, high-protein, low-calorie and low-calorie high-protein diets improved anxiety. Low-calorie high-protein diet could increase the recognition index as a representative of working memory. Clinical and experimental studies found that a high-fat diet impairs cognitive function and increases anxiety in rodents and healthy human [35][36][37][38]. In which these findings are in agreement with our data that the high-fat diet induces anxiety-like behavior in aged rats. On the other hand, calorie restriction is one of the most focused dietary patterns to improve cognition in the elderly [39,40] and current results established low-calorie high-protein diet had more influence on memory compared with lowcalorie regular food. Anti-aging protein Klotho and its co-receptor, FGF23 were boosted in the hippocampus and prefrontal cortex when rats were treated with high-protein, low-calorie and low-calorie high-protein diets confirming the role of the dietary protocol against anti-aging proteins [41]. The role of different diets on aging has been established and the Klotho signaling has been represented as an essential factor against aging [42]. Rios et al. [43] indicated that the high-fat diet reduced Klotho and FGF23 in the kidney. A clinical study showed the high-protein diet could improve Klotho and FGF21 serum level [44]; also, the low carbohydrate high-protein diet could induce FGF signaling and neuroprotection [45]. Data displayed low-calorie high-protein diet has a more influential role in klotho-α expression in the prefrontal cortex; these findings propose low-calorie high-protein diet is one of the promising candidates against neurodegeneration.
Also, we established Klotho-α level was increased in the low-calorie, high-protein and low-calorie high-protein diets in the serum, which confirmed Klotho-α circulate in the whole body and act their anti-aging function in many tissues such as the liver [46]. To find the precise role of diets on brain function, we evaluated the caspase-3 and some factors involved in neuronal plasticity in the hippocampus and prefrontal cortex. Data revealed caspase-3 level was increased during aging in the hippocampus, confirming the hippocampus's vulnerability rather than the prefrontal cortex [47]; moreover, high-fat diet increased caspase-3 level and low-calorie and low-calorie high-protein diets inhibited the caspase-3 activity in both hippocampus and prefrontal cortex. Neuronal plasticity factors such as TGF-β1, c-fos and BDNF were decreased during aging and high-fat diet intensifies this reduction in the brain; in contrast, high-protein, low-calorie and low-calorie high-protein diets could enhance these factors suggesting the role of these diets against neuronal degeneration and cognition [48,49]. Furthermore, data analysis indicated low-calorie high-protein diet has a more beneficial role in cognitive-related factors than a high-protein diet, confirmed by the in vivo and clinical studies [50,51].
According to our data, low-calorie and low-calorie high-protein diets are the most protective diet, so we chose low-calorie high-protein diet for the second step of the experimental procedure, and Klotho inhibitor, d-saccharic acid 1,4-lactone has been used. Inhibition of Klotho decreased c-fos and BDNF level in the low-calorie high-protein received rats, which established the vital role of different diets against neurodegeneration. FGF/ Klotho has a potential effect on c-fos in the kidney and liver [52,53], and these results showed that Klotho/c-fos signaling is presented in the brain. Inhibition of Klotho reduces BDNF and c-fos neuronal level, which shows BDNF and c-fos are downstream signal transduction for Klotho and might enhance working memory in the old rats. Previous studies well establish the role of BDNF in the Klotho signaling, and it has been shown that mutation of Klotho decreases BDNF level and also inhibition of BDNF could impair cognitive dysfunction in the Klotho mutant mice [54]. It is noteworthy that the d-saccharic acid 1,4-lactone inhibits both Klotho-α and -β, so the future study is suggested to be done to explore the effect Klotho-α inhibition on diets by using a specific inhibitor of Klotho-α or Klotho-α gene knockout.

Conclusion
Data found diets such as high-protein, low-calorie, lowcalorie high-protein diets increased Klotho/FGF23 signaling in the brain, and consequently, Klotho could enhance neuronal plasticity factors and enhance memory and anxiety. The result of the current study can be used by clinical practice and helps understand the Table 4 Correlation between Klotho protein and FGF23 protein, TGF-β protein, caspase-3 protein, c-fos and BDNF gene expression (n = 3 and technical replication is 3) P value less than 0.05 considered significant anxiety-like disorder's mechanisms and treatments. Also, many shreds of evidence support that the ketogenic diet, which contains low-carbohydrate high-protein or calorie-restricted diets containing low-carbohydrate, could improve cognition in healthy or aged humans. On the other hand, the study was conducted in male rats, and it might not be easy to extrapolate the results to female rats. Taken together, data confirmed that low-calorie and lowcalorie high-protein diets could potentially reduce aging and cognitive deficits by elevation of Klotho (Graphical abstract).