Effects of metformin and exercise on Alzheimer's model induced by D-galactose in the rats: A behavioral, histological and BDNF gene study Authors: Tahereh Varmazyar, Nasrin Heidariyeh, Shima Ababzadeh Published: 2026 ABSTRACT The objective of the current research is to investigate the cognitive performance of male Alzheimer's disease rats in the context of the combined effects of exercise and metformin. The rats were divided into eight different groups. The Morris water maze (MWM) and the open field test (OFT) were employed to assess the cognitive functionality and motor activity of rats, respectively. The hippocampus of rats was subjected to histological evaluation, and expression of brain-derived neurotrophic factor (BDNF) was measured using the real-time polymerase chain reaction (RT-PCR) method. On concluding day of training, the d-gal-met100-exe group (D-galactose plus optimized 100mg/kg of metformin along with exercise) exhibited a diminished duration in the target quadrant (P < 0.0001) and covered a shorter distance to find the platform (P = 0.0004). On the day of the test, the distance covered in the target quadrant decreased in the d-gal group that received 120mg/kg of D-galactose (P = 0.0186) and significantly increased in the d-gal-met100-exe group compared to the d-gal group (P = 0.0002). The OFT indicated no substantial variations in distance or overall speed across the groups, suggesting that metformin did not affect motor activity. Based on the histology findings, in the d-gal-met100-exe group, fewer pyramidal cells led to less neuronal damage (P < 0.0001). In contrast, the BDNF gene in the hippocampus of the aforementioned group rose considerably as compared to the d-gal group (P = 0.0002). In general, the combined administration of metformin and exercise had a greater effect on enhancing memory and learning in Alzheimer's rats. Introduction Alzheimer's disease is a complex neurodegenerative condition predominantly caused by age and genetic abnormalities [1,2]. Its prevalence rises exponentially beyond the age of 65 [3]. Previous study has linked the etiology of Alzheimer's disease to nerve cell degeneration and hippocampus neuron loss [4]. Brain-derived neurotropic factor (BDNF) is a neurotropic factor that promotes neuronal survival and growth. It plays an important role in memory formation and synaptic plasticity and is present in highly adaptable brain areas such as the hippocampus and cortex. Supplementary evidence suggests that the gradual decline of cognition and memory contributes to Alzheimer's disease, since low levels of BDNF are caused by progressive atrophy of neurons and synapses in the brain [5,6]. BDNF has a role in neural plasticity and impacts processes like long-term potentiation (LTP), which are essential for memory and learning [7,8]. Learning is a brain process that causes behavioral change via the acquisition of experience and memory. Learning is the process of storing and retrieving learned knowledge [9]. Among the multiple brain regions, the hippocampus and prefrontal cortex are thought to be key components of memory consolidation and organization [10]. Memory and other cognitive function impairments have been connected to hippocampal changes during the disease's early stages [11]. The Morris water maze (MWM) behavioral test is an important tool for assessing hippocampal function [12]. D-galactose (d-gal) is an endogenous molecule that is normally converted to glucose by the enzymes d-galactokinase and galactose-1-phosphate uridyltransferase [13]. When d-gal levels exceed the capacity of the two previously mentioned enzymes, the excess cannot be metabolized, resulting in d-gal accumulation, which can be converted into galactitol by aldose reductase or oxidized to produce galactonate, causing neurological dysfunction [14]. D-gal given to rats for 6 to 10 weeks causes neuronal death, which leads to age-related changes in the brain and the formation of an Alzheimer's disease model [15,16]. A series of studies indicated that increased inflammatory markers (IL-6, TNF-α), impaired spatial memory, and mild Aβ/Tau accumulation in the mouse hippocampus partially reproduce about 40–60% of the behavioral features of Alzheimer's disease [17]. Currently, metformin (1,1-dimethyl biguanide hydrochloride) is the most often given medicine for type 2 diabetes [18]. Metformin has been shown in studies to easily cross the blood-brain barrier [19] and reduce glucose production by reducing neuronal cell death via AMP-activated protein kinase (AMPK) stimulation pathways [20]. This topic relates to memory and learning impairments and is useful for improving laboratory mice models [21]. Physical exercise is widely known to help prevent cognitive decline and ameliorate Alzheimer's disease [22]. Exercise has been demonstrated in laboratory mouse studies to activate the hippocampus, resulting in an increase in neurogenesis inside hippocampal cells [23]. Furthermore, exercise increases BDNF gene expression, which controls exercise's effects on brain plasticity and, as a result, lowers the frequency of learning and memory impairments [24,25]. This research used the MWM behavioral test to investigate the synergistic effects of metformin and exercise on the improvement of memory and learning in male rats with Alzheimer's disease caused by d-gal. In addition, open field test (OFT) was used to examine locomotor activity of treated rats. This study was carried out by measuring BDNF gene expression using Real-time polymerase chain reaction (RT-PCR) and histological changes in the Cornu Ammonis (CA1) area of the hippocampus with Nissl staining. Materials and methods Animal The Pasteur Institute of Iran furnished 64 adult male Westar rats (weighing 250 ± 30 g) for the investigation. During the experiment, rats were housed at a temperature of 25 ± 1 °C, constant humidity, and a 12:12 h's light/dark cycle (lights on at 7:00 a.m. and off at 7:00 p.m.). Food and drink were always available. All experimental techniques were carried out in accordance with the recommendations of the Laboratory Animal Ethics Committee at Islamic Azad University's Qom branch, using the permitted code IR. IAU.QOM.REC.1401.108. Medications Metformin was acquired from Exir Pharmaceutical Company in Borujerd, Lorestan Province, Iran. A purchase of d-gal was made from Merck Pharmaceuticals (Germany). Metformin was administered intraperitoneally to the rats at dosages of 50 mg/kg, 100 mg/kg, and 200 mg/kg body weight, diluted in sterile saline (0.9% NaCl) [26]. The Alzheimer's disease model was established for a period of 6 weeks by administering an intraperitoneal injection of d-gal at a dosage of 120 mg/kg, which was dissolved in sterile saline [27]. The purity of all substances was of the highest standard. Study design The rats were divided into eight different groups of eight after a week of acclimation to the experimental environment. The following is how the groups were defined: (1) the intact group, which consisted of unaltered animals that received no intervention; (2) the control group, which received 0.5 cc of 0.9% saline for six weeks (saline); (3) the group that received 120 mg/kg of d-gal in 0.9% saline to simulate Alzheimer's disease (d-gal); (4) the group that received 120 mg/kg of d-gal in addition to 50 mg/kg of metformin (d-gal-met50); (5) the group that received 120 mg/kg of d-gal combined with 100 mg/kg of metformin (d-gal-met100); (6) the group that received 120 mg/kg of d-gal combined with 200 mg/kg of metformin (d-gal-met200); (7) the group that received 120 mg/kg of d-gal along with exercise (d-gal-exe); and (8) the group that received 120 mg/kg d-gal plus optimal dosage of 100 mg/kg of metformin in addition to exercise (d-gal-met100-exe). All groups 4, 5, 6, 7, and 8 received d-gal injections throughout a 42-day period. Started on the 22nd day after d-gal injection, the groups of 4, 5 and 6 was given 50, 100 and 200 mg/kg of metformin for 21 days, respectively. Starting on the 22nd day after the d-gal injection, the d-gal-exe group engaged in 20–25 min of exercise in every day except of two days each week. After the determination of 100 mg/kg of metformin injection as an optimal dosage, the d-gal-met100-exe group had the same exercise and metformin injection regimens as the other groups. In particular, exercise was done for 20 to 25 min every day, save for two days a week, and metformin at a dose of 100 mg/kg was given once daily for 21 days, starting on the 22nd day following the d-gal injection. After medication administration and exercise, the OFT and the MWM test were used to assess motor activity and spatial memory. OFT Initially, each rat was placed in the center of the OFT's box, which is a black and open metal box measuring 100 cm in length, 100 cm in width, and 40 cm in height, with the floor divided into 25 rectangles by marked lines. A digital camera was placed above the box to record the rats' behavior, and after each experiment, the receptacle was carefully cleaned with a cotton cloth saturated with 70% ethanol. The total distance traveled and the overall paces were recorded and analyzed [28]. MWM behavioral test The most widely used and standardized behavioral test for assessing an animal's spatial learning and memory is the MWM test [29]. Rats need to be able to swim in a pool and learn to find a hidden platform in the water. During the training days, the rats learn the position of the platform via a process known as spatial learning [30]. On the test day without the platform, the rats' spatial memory is evaluated by measuring the amount of time they spend swimming in the target quadrant and the distance they cover in that region [12]. The MWM test was carried out in a circular pool with a diameter of 160 cm and a height of 50 cm that was split into four quadrants. A pentagon, square, circle, and triangle were among the forms used to decorate the pool's perimeter. The water level in the pool was set at 27 cm, and the temperature was controlled between 23 and 24 °C. A 12-cm-diameter platform was submerged 1.2 cm below the water's surface. The rats' swimming time was set at 60 s, and a video camera was placed above the tank to record their movements and duration throughout the experiment. The testing chamber had no direct light and was acoustically insulated. Days of acquisition (learning phase) All rats were sufficiently equipped for instruction one day before the learning phase commenced. During this phase, each rat underwent a total of twelve trials, with four trials being administered on each day for three consecutive days [24,31]. The rat was liberated from one of the labyrinth's quadrants and confronted the maze wall in each trial. After 60 s or when the rat reached the platform, each trial was terminated. The mouse was carefully guided to the platform and allowed to remain for 15 s if it did not uncover it within the designated time. Recall test (memory retrieval phase) On the fourth day, known as test day, the platform was removed from the pool, enabling the rat to swim freely for 1 min [30]. The time spent in the target zone was monitored and analyzed to ascertain spatial reference memory retention. After exiting the water, the rats were hand-dried with a cotton towel. Exercise protocol One week before the start date, the rats were introduced to and acclimated to training conditions at a speed of 5 to 8 m per minute for 10 to 15 min. The exercising groups used a rat treadmill for three weeks. The training schedule included jogging on the treadmill at 12 m per minute in the first week, 14 m per minute in the second week, and 16 m per minute in the third week. The d-gal-exe group engaged in 20–25 min of exercise in every day except of two days each week [32]. The selection of this protocol, specifically the type of exercise and training modality was aligned with the study's objective and was, as far as possible, standardized and repeatable. Furthermore, before the exercise protocol began, rats were placed on the treadmill under specific conditions for a set duration. The groups undergoing exercise treatment were placed in the treadmill device, initially without movement. This process could independently reduce the initial stress responses caused by treadmill exercise. Histological studies and gene expression in Alzheimer's rats After the behavioral tests were completed, the rats were sedated with chloroform for 24 h. Four rats' brains were extracted for hippocampus tissue studies in each group using the Paxinos & Watson mouse brain atlas and fixed in 10% formalin. The hippocampus of the other four rats in the group were isolated on ice, placed in sterile cryotubes, frozen in liquid nitrogen, and then stored at −80 °C. Ultimately, the hippocampus was used to quantify BDNF gene expression [33]. Histological studies and nissl staining Brain sections that were preserved in 10% formalin for one week were subjected to paraffin embedding. The brain tissues were then affixed on gelatin-coated transparencies after being cut to a thickness of 4 μm using a rotary crotome (Leica RM2235, Germany), and were followed by Nissl staining method. The Nissl staining (utilizing toluidine blue) was employed to evaluate the severity of injury and to quantify the number of healthy and degenerated neurons in the CA1 region of the hippocampus [34]. Four sections (anterior 1/4, second 1/4, third 1/4, and posterior 1/4) were prepared from each sample. Cell counting was conducted within four 5-μm interval squares per section using Image J software, followed by observation under an Olympus microscope. The average count was taken as the sample's cell number. Study of BDNF gene expression BDNF gene expression was estimated using hippocampal tissue that had been preserved in cryotubes and stored at −80 °C, as previously mentioned. In accordance with the appropriate instructions, the TRIZOL reagent and the One Step-RNA Reagent from Bio Basic Corporation were used to extract hippocampal RNA from the brain tissue [35]. The quality of RNA was assessed using a Nano Drop spectrophotometer (Nano Drop One, Thermo Scientific, USA) at a wavelength of 260 to 280 nm. cDNA was synthesized using the RNA that was obtained. This was accomplished by utilizing the YT4500 cDNA Synthesis Kit from Yekta Tajhiz Azma (Iran). The primer characteristics for the BDNF gene, which serves as the target gene, and β-actin, which serves as the internal control gene, were obtained using the NCBI Primer-BLAST website. The forward and reverse primers were created using Gene Runner software. RT-PCR was employed to evaluate the expression of the BDNF gene subsequently to the extraction of RNA from the cells and the synthesis of cDNA. This procedure was conducted using the One Plus device (Applied Biosystems) and the ensuing temperature-time program: Initially, the thermo-cycler program was set up to conduct a single cycle of cDNA denaturation at 95 °C for 15 min. This was followed by 40 cycles, which consisted of DNA strand synthesis at 72 °C for 40 s, annealing for 30 s, and denaturation at 95 °C for 20 s. The melt curve was determined from 60 °C to 95 °C, and all reactions for all genes were performed in triplicate. The melt curve analysis was employed to confirm the specificity of PCR amplification [36,37]. Following the reaction, the Cycle Threshold (Ct) was compared to analyze the RT-PCR data. The calculated discrepancies in Ct values between the control samples (untreated samples) and the tested samples (treated samples) were analyzed in this investigation. The fold change, which is the ratio of the target gene to the reference gene (β-actin), was determined using Equation (1) (eq. (1)) [38]. This method was employed to evaluate the target gene expression ratio between the experimental and control groups.eq. (1)Fold change: BDNF relative expression = 2–ΔΔCteq. (2)ΔCt = Cttarget - Ctreferenceeq. (3)ΔΔCt = ΔCttestsample - ΔCtcontrolsample The experiment was conducted three times to ascertain the average value. The standard gene was employed as an internal control, and the change in gene expression was compared to it. This gene is typically a maintenance gene, and β-actin was implemented in this investigation. The attributes of the BDNF and β-actin loci are summarized in Table 1. The re-amplification of the BDNF and β-actin genes has been effectively completed, as indicated by the primers' characteristics.Table 1Characteristics of BDNF and β-actin genes.Table 1GenePrimer typeSequenceBDNFForwardCCTCTGCTCTTTCTGCTGGAReverseTATCTGCCGCTGTGACCCβ-actinForwardACGTTGACATCCGTAAAGACCReverseGCCACCAATCCACACAGAGT Table 1Characteristics of BDNF and β-actin genes.Table 1GenePrimer typeSequenceBDNFForwardCCTCTGCTCTTTCTGCTGGAReverseTATCTGCCGCTGTGACCCβ-actinForwardACGTTGACATCCGTAAAGACCReverseGCCACCAATCCACACAGAGT Characteristics of BDNF and β-actin genes. Statistical data analysis All statistical analyses were conducted using GraphPad Prism version 8.2.1 Inc., San Diego, CA, USA [39]. EthoVision XT was the most frequently used video tracking software for the purpose of monitoring and analyzing the behavior, movement, and activity of each animal, and it was used for the OFT and MWM test. In addition to one-way and two-way analysis of variance (ANOVA), Tukey's post hoc test was implemented to evaluate additional data. The data are represented as the mean ± standard error of the mean (Mean ± SEM). A P-value of less than 0.05 (P < 0.05) was used to define statistical significance. Results Determination of the effective dosage of metformin In order to examine the impact of three concentrations of metformin, 50, 100 and 200 mg/kg, on Alzheimer's rats, the OFT and MWM test were initially implemented. OFT Based on the results depicted in Fig. 1, the all of d-gal-met50, d-gal-met100 and d-gal-met200 groups have no significant changes in distance traveled and speed when compared to the d-gal group.Fig. 1Overall movement speed (A) and overall distance traveled (B) in the groups with different dosages of metformin and the groups of intact, saline, and d-gal assessed by the OFT. The data are presented as the Mean ± SEM. The data was analyzed using a one-way ANOVA with Tukey's post hoc test. Overall movement speed (A) and overall distance traveled (B) in the groups with different dosages of metformin and the groups of intact, saline, and d-gal assessed by the OFT. The data are presented as the Mean ± SEM. The data was analyzed using a one-way ANOVA with Tukey's post hoc test. MWM The results showed that the met100 treatment group spent significantly more time in the target quartile than the d-gal group (P = 0.0025), but no significant difference was seen among the d-gal-met50 and d-gal-met200 groups compared to d-gal group (see Fig. 2A). The distance traveled to locate the platform in the target quadrant was significantly greater in the group treated with 100 mg/kg metformin than in the d-gal group on the test day (P = 0.0038). In contrast, the groups given 50 and 200 mg/kg of metformin showed no significant alterations from the d-gal group (Fig. 2B).Fig. 2The MWM behavioral test is employed to ascertain the optimal dosage of metformin. A: The duration of time spent in the target quartile on the test day, B: The distance traveled to locate the platform in the target quartile on the test day, C: Number of entries by rats into the target quartile. Data are expressed as the Mean ± SEM. The data was analyzed using a one-way ANOVA with Tukey's post hoc test. The P-value comparisons for various categories are presented: ##P < 0.01 in comparison to the d-gal group; ∗P < 0.05 in comparison to the saline; ns: no significant difference. The MWM behavioral test is employed to ascertain the optimal dosage of metformin. A: The duration of time spent in the target quartile on the test day, B: The distance traveled to locate the platform in the target quartile on the test day, C: Number of entries by rats into the target quartile. Data are expressed as the Mean ± SEM. The data was analyzed using a one-way ANOVA with Tukey's post hoc test. The P-value comparisons for various categories are presented: ##P < 0.01 in comparison to the d-gal group; ∗P < 0.05 in comparison to the saline; ns: no significant difference. Furthermore, the met100 mg/kg group had a significantly higher number of rats entering the target quartile than the d-gal group (P = 0.0017). Nevertheless, there was no noticeable difference among the met50 mg/kg-treated, met200 mg/kg-treated and d-gal groups (Fig. 2C). As a result, the efficacy dosage of metformin was found to be 100 mg/kg. Overall, the results reported in Fig. 2 suggest that the 100 mg/kg dose of metformin led to significantly longer duration and distance traveled in the target quadrant compared to the 50 mg/kg and 200 mg/kg doses. Furthermore, a greater number of rats treated with 100 mg/kg metformin entered the target quadrant, indicating improved memory performance. In contrast, one of effect of administering the 200 mg/kg dose included the mortality of two rats. Our investigations were continued for the intact, saline, d-gal, d-gal-met100, d-gal-exe, and d-gal-met100-exe groups and the results were reported as follows. The effect of metformin and exercise on motor performance (OFT) Prior to the memory evaluation, an OFT was delivered to determine the effect of metformin and exercise on rat motor performance. This evaluation examined the rats' general speed and total distance traveled. The findings showed that the d-gal group and the saline group, as well as the d-gal-met100, d-gal-exe, and d-gal-met100-exe groups, had no significant changes in distance or total pace of movement when compared to the d-gal group (Fig. 3A and B).Fig. 3Overall movement speed (A) and overall distance traveled (B) in different study groups assessed by the OFT. The data are presented as the Mean ± SEM. The data was analyzed using a one-way ANOVA with Tukey's post hoc test. Overall movement speed (A) and overall distance traveled (B) in different study groups assessed by the OFT. The data are presented as the Mean ± SEM. The data was analyzed using a one-way ANOVA with Tukey's post hoc test. In addition to motor activity, this test also examines stress. The number of entries into the center zone in the OFT can be considered as one of the assessments for stress [40]. In this research work, the number of entries into the center was relatively high, which suggests negligible stress levels. Since the objective of the current study was to examine exploratory activities in the OFT, any discussions related to stress have been omitted. The effect of metformin and exercise on the behavior (MWM) during acquisition and recall days The MWM behavioral evaluation is separated into two phases: acquisition (learning) and retention (memory recall). The distance traveled, time spent swimming in the target quadrant, and swimming speed were all measured in this test. To do statistical analysis on the findings, a two-way ANOVA was used for the acquisition days and a one-way ANOVA for the retention day. Distance traveled during the acquisition days The d-gal group went much further than the saline, as demonstrated by the findings presented at Fig. 4. This impact was substantial (P = 0.0097). On the third day of training, the Alzheimer's-induced rats traveled significantly less than the d-gal group after receiving a 100 mg/kg injection of metformin (d-gal-met100 group, P = 0.0025). Furthermore, Alzheimer's-induced rats group that was exercised (d-gal-exe group) showed a significant difference from the d-gal group on the third day of training (P = 0.0054). Furthermore, despite getting exercise and metformin injections, the Alzheimer's-induced rats group (d-gal-met100-exe) performed very well in selecting the target platform. The distance traversed by this group was substantially reduced, demonstrating the most substantive difference in comparison to the d-gal group (P = 0.0004). This indicates that these rats' memory and learning capacities have been boosted.Fig. 4Distance traveled on different days of the learning phase for various groups as assessed by the MWM behavioral test. The data are presented as the Mean ± SEM. The data was analyzed using a two-way ANOVA with Tukey's post hoc test. The P-value comparisons for various categories are presented: ###P < 0.001, ##P < 0.01 and #P < 0.05 in comparison to the d-gal group; ∗P < 0.05 and ∗∗P < 0.01 in comparison to the saline; ns: no significant difference. Distance traveled on different days of the learning phase for various groups as assessed by the MWM behavioral test. The data are presented as the Mean ± SEM. The data was analyzed using a two-way ANOVA with Tukey's post hoc test. The P-value comparisons for various categories are presented: ###P < 0.001, ##P < 0.01 and #P < 0.05 in comparison to the d-gal group; ∗P < 0.05 and ∗∗P < 0.01 in comparison to the saline; ns: no significant difference. Swimming time on acquisition days The results indicate a significant difference between the d-gal group and the saline group throughout the learning process (P = 0.0024). The rats developed Alzheimer's-like symptoms after receiving a dosage of 120 mg/kg of d-gal, which led to an increase in time spent in the target quadrant. The two-way ANOVA found a significant difference in all acquisition days between Alzheimer's rats and the d-gal-exe, d-gal-met100, and d-gal-met100-exe groups. Hence, all derived P-values were significant. On the third day, all groups had P-values of 0.0020, 0.0011, and <0.0001, respectively. This means that the d-gal-met100-exe group performed much better than the other groups, since the three treatment groups were able to find the platform in the target quadrant in a shorter period of time. This shows that metformin and exercise work well together to enhance learning and memory in this group (Fig. 5).Fig. 5Swimming time in the target quadrant on different days of the learning phase in various groups as measured by the MWM behavioral test. The data are presented as the Mean ± SEM. The data was analyzed using a two-way ANOVA. The P-value comparisons for various categories are presented: #P < 0.05, ##P < 0.01, and ####P < 0.0001 in comparison to the d-gal group, and ∗P < 0.05 and ∗∗P < 0.01 in comparison to the saline; ns: no significant difference. Swimming time in the target quadrant on different days of the learning phase in various groups as measured by the MWM behavioral test. The data are presented as the Mean ± SEM. The data was analyzed using a two-way ANOVA. The P-value comparisons for various categories are presented: #P < 0.05, ##P < 0.01, and ####P < 0.0001 in comparison to the d-gal group, and ∗P < 0.05 and ∗∗P < 0.01 in comparison to the saline; ns: no significant difference. Swimming speed on acquisition days During the training period, the swimming speed of rats in different training groups was measured. Fig. 6 shows that there was no significant difference in average swimming speed between the saline group and the other treatment groups. This shows that metformin and exercise have not any effect on their movement behavior.Fig. 6Swimming speed in the target quadrant on different days of the learning phase in various groups, assessed by the MWM behavioral test. The data are presented as the Mean ± SEM. The data was analyzed using a two-way ANOVA with Tukey's post hoc test. Swimming speed in the target quadrant on different days of the learning phase in various groups, assessed by the MWM behavioral test. The data are presented as the Mean ± SEM. The data was analyzed using a two-way ANOVA with Tukey's post hoc test. Distance traveled for finding the platform on the recall day The findings show that following an intraperitoneal injection of d-gal at a dosage of 120 mg/kg, the distance traveled in the target quadrant reduced significantly compared to the saline group (P = 0.0186). The Alzheimer's rats showed a significant increase in the distance traveled in the target quadrant with the administration of metformin at 100 mg/kg (P = 0.0038). Furthermore, when Alzheimer's rats were exercised, they performed better at identifying the target platform (P = 0.0050). The P-value for the group that underwent exercise simultaneously with a 100 mg/kg metformin injection was 0.0002 (P = 0.0002), according to one-way ANOVA. In contrast to the Alzheimer's rats, they went the greatest distance in the target quadrant to find the platform, suggesting that concurrent metformin and exercise had a positive impact on Alzheimer's disease reduction in the examined rats (Fig. 7).Fig. 7Distance traveled on the test day in different groups measured by the MWM behavioral test. The data are presented as the Mean ± SEM. The data was analyzed using a one-way ANOVA. The P-value comparisons for various categories are presented: ##P < 0.01 and ###P < 0.001 in comparison to the d-gal group, and ∗P < 0.05 in comparison to the saline; ns: no significant difference. Distance traveled on the test day in different groups measured by the MWM behavioral test. The data are presented as the Mean ± SEM. The data was analyzed using a one-way ANOVA. The P-value comparisons for various categories are presented: ##P < 0.01 and ###P < 0.001 in comparison to the d-gal group, and ∗P < 0.05 in comparison to the saline; ns: no significant difference. Time spent swimming in the target quadrant on the recall day Fig. 8 depicts the swimming time of the diseased group (d-gal) and the therapy groups, indicating that the time spent swimming in the target quadrant increased across all treatment groups. The group that received 120 mg/kg of d-gal spent significantly less time in the target quadrant than the saline group on the test day when the platform was removed (P = 0.0146). The results suggest that the Alzheimer's-induced group had poorer spatial memory. Fig. 8 also shows a significant difference between the d-gal group and the d-gal-met100 and d-gal-exe groups (P-values of 0.0025 and 0.0052, respectively). The sick group that got 100 mg/kg metformin and exercised concurrently (d-gal-met100-exe) spent the greatest time in the target quadrant looking for the platform, suggesting a very significant difference (P = 0.0002).Fig. 8Swimming time in the target quadrant on the test day for various groups using the MWM behavioral test. The data are presented as the Mean ± SEM. The data was analyzed using a one-way ANOVA. The P-value comparisons for various categories are presented: ##P < 0.01 and ###P < 0.001 in comparison to the d-gal group, ∗P < 0.05 in comparison to the saline; ns: no significant difference. Swimming time in the target quadrant on the test day for various groups using the MWM behavioral test. The data are presented as the Mean ± SEM. The data was analyzed using a one-way ANOVA. The P-value comparisons for various categories are presented: ##P < 0.01 and ###P < 0.001 in comparison to the d-gal group, ∗P < 0.05 in comparison to the saline; ns: no significant difference. Fig. 9 displays the swimming path image for the all study groups (control and treated). The rats in the d-gal-met100-exe group traveled the greatest distance in the target quadrant, indicating they spent more time in that area. These images confirm the findings presented in Fig. 7, Fig. 8.Fig. 9Swimming path images for the all groups along with recorded data relevant to each zone on the recall day in the MWM test for one rat selected randomly in each group. Swimming path images for the all groups along with recorded data relevant to each zone on the recall day in the MWM test for one rat selected randomly in each group. Swimming speed on the recall day On test day, the hidden platform was discovered by monitoring the swimming speed of rats in different groups. Fig. 10 shows that there was no statistically significant difference in mean swimming speed between the control (saline) and other treatment groups.Fig. 10Swimming speed to find the hidden platform on the test day across different groups using the MWM behavioral test. The data are presented as the Mean ± SEM. The data was analyzed using a one-way ANOVA. Swimming speed to find the hidden platform on the test day across different groups using the MWM behavioral test. The data are presented as the Mean ± SEM. The data was analyzed using a one-way ANOVA. The effect of metformin and exercise on the number of damaged cells (histology) Statistics studies Significant differences in the number of damaged pyramidal neurons in the CA1 area of the hippocampus were seen between rats in the saline and d-gal groups (P < 0.0001), d-gal and d-gal-met100-exe groups (P < 0.0001), d-gal and d-gal-exe groups (P = 0.0009), and d-gal and d-gal-met100 groups (P = 0.0002). These findings suggest that the d-gal group had memory impairment and pyramidal hippocampus cell apoptosis, but the treated rats had less hippocampal cell loss, showing that exercise and metformin had a favorable impact on these cells (Fig. 11).Fig. 11The number of damaged neurons in the CA1 region of the hippocampus across different groups. The data are presented as the Mean ± SEM. The data was analyzed using a one-way ANOVA. The P-value comparisons for various categories are presented. ###P < 0.001 and ####P < 0.0001 in comparison to the d-gal group, and ∗∗∗∗P < 0.0001 in comparison to the saline; ns: no significant difference. The number of damaged neurons in the CA1 region of the hippocampus across different groups. The data are presented as the Mean ± SEM. The data was analyzed using a one-way ANOVA. The P-value comparisons for various categories are presented. ###P < 0.001 and ####P < 0.0001 in comparison to the d-gal group, and ∗∗∗∗P < 0.0001 in comparison to the saline; ns: no significant difference. Pictorial studies Fig. 12A depicts the CA1 area of the hippocampus. The d-gal group had more deteriorated cells than the other treated groups, as seen in Fig. 12D. The d-gal group had a significantly lower number of healthy neurons, but the d-gal-exe, d-gal-met100, and d-gal-met100-exe groups had a higher number of healthy neurons than the diseased group.Fig. 12Nissl staining was used in the hippocampus of the d-gal group and all treated groups. A: The CA1 region of the hippocampus (4× magnification), B, C, D, E, F, and G: The CA1 region of the hippocampus was examined in the intact, saline, d-gal, d-gal-exe, d-gal-met100, and d-gal-met100-exe groups, respectively (40× magnification). Healthy pyramidal neurons with visible nuclei and distinct nucleoplasm are indicated by black arrows, while neurons in the CA1 region that have been destroyed are indicated by red arrows. Nissl staining was used in the hippocampus of the d-gal group and all treated groups. A: The CA1 region of the hippocampus (4× magnification), B, C, D, E, F, and G: The CA1 region of the hippocampus was examined in the intact, saline, d-gal, d-gal-exe, d-gal-met100, and d-gal-met100-exe groups, respectively (40× magnification). Healthy pyramidal neurons with visible nuclei and distinct nucleoplasm are indicated by black arrows, while neurons in the CA1 region that have been destroyed are indicated by red arrows. The effect of metformin and exercise on BDNF gene expression in the hippocampus Equation (1) is used to compute gene expression, as shown in section 2.7.2. This research used RT-PCR to explore the effect of exercise and metformin on BDNF gene expression in rat hippocampus. β-actin was used as an internal reference gene. The findings showed significant differences between the d-gal and control (saline) groups (P < 0.0001), the d-gal and d-gal-met100-exe groups (P = 0.0002), the d-gal and d-gal-exe groups (P = 0.0263), and the d-gal and d-gal-met100 groups (P = 0.0172). The results have been reported in Fig. 13. Metformin and exercise have a positive impact on the hippocampus of treated rats, as shown by higher expression of the BDNF gene, which leads to improved memory.Fig. 13Comparison of BDNF expression in the hippocampus of rats between the diseased group and the treatment groups The data are presented as the Mean ± SEM. The data was analyzed using a one-way ANOVA. The P-value comparisons for various categories are presented. #P < 0.05 and ###P < 0.001 compared to the d-gal group, and ∗∗∗∗P < 0.0001 compared to the saline; ns: no significant difference. Comparison of BDNF expression in the hippocampus of rats between the diseased group and the treatment groups The data are presented as the Mean ± SEM. The data was analyzed using a one-way ANOVA. The P-value comparisons for various categories are presented. #P < 0.05 and ###P < 0.001 compared to the d-gal group, and ∗∗∗∗P < 0.0001 compared to the saline; ns: no significant difference. Discussion The present work investigates the protective effects of exercise and metformin on memory and mRNA expression for BDNF in the hippocampus of d-gal-induced Alzheimer's rats. Alzheimer's disease causes deterioration of cognitive function and memory. The d-gal is one of the variables contributing to this condition [41]. Previous studies have shown that increased levels of d-gal promote the formation of advanced glycation end products (AGEs). Abnormally high quantities of these products may boost caspase-3 expression while decreasing AMPK and Phosphoinositide 3-kinase (PI3K) expression. In conclusion, the reduction of these genes induces the degenerative processes of neurological illnesses, resulting in a fall in spatial learning and memory [13,42] and lower BDNF levels [13,43]. The MWM test revealed that the group that took d-gal in the current research had memory impairment. The results of this group were similar with prior research [44]. In particular, they spent more time scanning for the platform in the target quadrant on the final training day and less time on the recall test day than the control group. Metformin has been shown in previous studies to have neuroprotective effects in d-gal-induced neurological diseases [45,46] and to improve brain function by regulating brain metabolism and homeostasis [47]. Sarkaki and colleagues revealed metformin's neuroprotective effects on memory improvement in the OFT [40]. Furthermore, Wang and colleagues found that metformin enhances learning and memory in the MWM test and stimulates neurogenesis [48]. The motor behavior of Alzheimer's rats was not favorably influenced by metformin at the dosages of 50 and 200 mg/kg, according to the findings of this research. As a result, their memory was badly impaired. In contrast, the rats tested showed sufficient nervous system protection and increased memory at the optimum dosage of 100 mg/kg. Thangthaeng and colleagues also found that higher doses had an unfavorable effect on male rat performance. The researchers discovered that dosages of 219 and 297 mg of metformin per kg had a detrimental influence on spatial memory recovery and did not enhance learning processes [49]. In addition, our histological studies revealed that metformin at the optimum dosage of 100 mg/kg improved the quantity and survival of CA1 pyramidal neurons in the hippocampus [50]. Metformin's beneficial effect is linked to its capacity to improve neuronal survival and the production of neuroprotective genes via signaling pathways such as AMPK/CREB (cAMP-response element-binding protein) [51]. Previous studies showed that metformin treatment increased the amount of synaptophysin (a presynaptic membrane protein) and improved hippocampus tissue shape when compared to the control group. This activated the AMPK/BDNF/PI3K signaling pathway, increasing neurogenesis and spatial memory formation [43,52]. In the present study, rats treated with metformin exhibited less memory impairment on the final day of training in the MWM test than the d-gal group, and they spent less time in the target quadrant. Furthermore, on the test day, more time was spent in the target quadrant compared to the d-gal group, indicating that metformin improved memory. Our histopathological findings revealed neuronal loss in the d-gal group's hippocampus, while metformin protected hippocampal neurons. Metformin reduced neuronal damage in the hippocampus and crossed the blood-brain barrier, hence preventing d-gal-induced memory impairment. Furthermore, the results of the OFT in this investigation, which refer to the total distance traveled and overall movement speed, are comparable with the findings of Fang Wang and colleagues, indicating that metformin has no effect on the motor activity of treated mice [53]. Pak-Hashemi and colleagues have also confirmed this finding [54]. Exercise protects the brain via increased hippocampus neurogenesis and plasticity. These pathways cause structural changes in the brain, including increased volume, surface area, and total capillary length in the cerebral cortex, as well as neuronal plasticity and presynaptic density recovery in the hippocampus [55,56]. Exercise may protect CA1 hippocampal pyramidal neurons by strengthening molecular pathways, increasing levels of BDNF gens, neurogenesis, and plasticity, and ultimately boosting both short- and long-term memory [55,57,58]. Previous study has demonstrated that low-intensity exercise improves spatial learning and memory in rats, but higher intensity exercise degrades these capacities. Exercise intensity affects spatial learning and memory via regulating BDNF gene expression and glutamate and N-methyl-d-aspartate receptor (NMDAR) activation. Optimal exercise intensity stimulates NMDAR activation in neurons via signaling pathway related to cAMP response element-binding protein (p-CREB)/BDNF/NMDAR in the hippocampus [56]. The present study investigated the positive impacts of exercise on memory and learning, motor skills, and BDNF gene expression in a rat model of Alzheimer's disease. [57]. A schematic of the mechanism of metformin and exercise on memory is presented in Fig. 14A and B. The findings of this study are consistent with those of Nigam et al., which showed that three weeks of exercise enhanced memory in models of Alzheimer's disease [59]. Furthermore, this study found that the number of necrotic CA1 pyramidal cells in the hippocampus was much lower in Alzheimer's groups that exercised. Also, the results indicate that exercise increased BDNF gene expression that linked to improvements in rats' behavioral and spatial learning, as well as memory. The MWM test confirmed the results, demonstrating that on test day, the d-gal-exe group spent more time than the d-gal group recognizing the platform. This study found that increased intake of d-gal caused memory impairment, which resulted in decreased expression of the BDNF gene, neuronal atrophy, and death. In compared to the d-gal-met100 and d-gal-exe groups, administering 100 mg of metformin with exercise in Alzheimer's-induced rats resulted in a large increase in the expression of this gene, leading to improvements in memory and learning. Furthermore, the average expression of the BDNF gene in the hippocampus of rats treated with exercise plus metformin was considerably higher than in the other experimental groups. The behavioral MWM test revealed that rats given metformin and exercise walked fewer distances during training days than the d-gal group. Furthermore, the amount of time spent in the target quadrant on test day rose dramatically. Histological examinations revealed that this group's hippocampus had a significant increase in the number of healthy pyramidal cells, indicating improved memory. The combination of exercise and metformin injections reduced biomarkers linked with Alzheimer's disease while also protecting hippocampus function against d-gal-induced memory loss. The OFT findings showed no noticeable variation in the speed or general movement of the different groups. As a result, the neuroprotective properties of metformin plus exercise may help Alzheimer's rats learn and remember better. According to the results of this study, using metformin in combination with physical activity is an innovative, simple, and cost-effective way to prevent cognitive impairment in the hippocampus and apoptosis induced by d-gal.Fig. 14Diagram of how metformin and exercise effects neurogenesis in neurons. Diagram of how metformin and exercise effects neurogenesis in neurons. Conclusion The purpose of this research was to investigate the effect of treadmill exercise and metformin on the amelioration of Alzheimer's disease in male rats by protecting CA1 pyramidal neurons in the hippocampus and boosting BDNF gene expression. Our findings supported the use of d-gal in Alzheimer's disease investigations in rat models. The MWM test showed that metformin at a dosage of 100 mg/kg plus exercise could improve cognition and memory deficits caused by d-gal. The d-gal-met100-exe group had the most significant impacts on all analyzed parameters of the treatment groups, with the exception of swimming speed in the MWM test and total distance traveled in the OFT, according to the data. Also, the current study found that administering metformin plus exercise significantly increased the expression of the BDNF gene in the hippocampus, resulting in improved learning and memory in Alzheimer's rats. Therefore, using metformin with exercise as an innovative and successful method can prevent and cure d-gal-induced Alzheimer's in animal models. Ethics approval statement All experimental protocols were strictly followed by the recommendations of the Laboratory Animal Ethics Committee at Islamic Azad University's Qom branch, using the permitted code IR. IAU.QOM.REC.1401.108 as outlined in the “Guide for the Care and Use of Laboratory Animals” and published by the National Institutes of Health. Authors’ contributions Tahereh Varmazyar: Study design, data collection, data analysis and discussion, Writing – original draft, submission preparation. Nasrin Heidariyeh & Shima Ababzadeh: Investigation, discussion of the methods, results and interpretation, review and editing. The authors have read and agreed to publish a version of the manuscript. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.