BRIEF REPORT
OBJECTIVE: Aging is a multifactorial and progressive phenomenon, exclusively leading to loss of cellular, molecular and physiological functionality. It is well established that overproduction of free radicals such as reactive oxygen species (ROS) leads to oxidative stress which results in senescence and aging-related disorders. Moreover, the brain neurons are more susceptible to oxidative stress due to the presence of high lipid content and higher oxygen consumption.
The aim of this study was to investigate the effects of Metformin (Met), an antidiabetic biguanide, on learning and memory functions of young and aged rats using different behavioral tasks and combining with the PCR gene expression method as well as additional histological evidence, we were able to achieve a very unique result.

METHODS: Wistar-Albino male rats were separated into four groups: young mice (<12months-old), young mice +metformin, aged mice (24 months old), aged mice+ metformin. Metformin (100mg/kg) was supplemented into drinking water for 8 weeks. Morris water maze (MWM) and passive avoidance (PA) tests were used to determine learning and memory functions. Locomotor activity (locomotor activity cabinet system) was measured with a computerized system.

Locomotor activity test
Because old age and compounds altering locomotor activity may give false-positive/negative effects in behavioral tests, an additional test was carried out with the specific aim of monitoring motor activity. The spontaneous locomotor activity of the animals was assessed by monitoring their activity in a locomotor activity cage. Locomotor activity was measured with a computerized system (40×40×35 cm box; May Commat, Ankara, Turkey). Total number of movements was measured for a 5-min period before the behavioral tests and is expressed as the sum of stereotypic, ambulatory, and vertical activity. In this study, the 24-month-old naturally aged rats were chosen, which do not have impaired locomotor activity or any neurological deficit, in order avoid false-positive/negative effects in behavioral tests.

Passive avoidance test
In this type of avoidance learning test, the animals were refraining from making measured response. A step-down variant passive avoidance apparatus was used (Ugo Basile model 7551, Italy). The apparatus (measuring 22×21×22 cm) consisted of 2 compartments: a light and dark compartment separated by a guillotine door. On Day 1 (training trial), the rats were placed individually into light compartment and allowed to explore the boxes to become aware of environment.
1. Pre-acquisition trial: After 30 s, the door between the 2 boxes was opened, and animal moved into dark compartment freely.

2. The acquisition (training) trial was conducted 15 min after pre-acquisition trial. Rats were placed in light compartment, after 30-s adaptation period, door between the compartments were opened. Having completely entered dark compartment, door was automatically closed, and an electric foot-shock (0.5 mA) of 3-s duration was delivered to animal via grid floor. The time taken to reenter dark compartment was recorded (training latency). Any animal failing to cross from the light to dark compartment within 300 s was discarded from the experiment. Animals were then removed from dark compartment and returned to their home cages. Between each training session, both chamber compartments were cleaned to remove any confounding olfactory cues.

3. Retention trial: Recall of inhibitory stimulus was evaluated at 24-h post-training by returning animals to light compartment and recording their latency to enter dark compartment (4 paws in). No foot-shock was applied in this trial. If animal did not enter dark compartment within 300-s, it was returned to its cage and a maximum latency of 300-s was recorded. This latency served as a measure of retention performance of step- down avoidance responses (retention latency).

Morris water maze test
The Morris water maze consisted of a circular pool (150-cm diameter) was filled with water (25°C) and rendered opaque by addition of small white pieces of plastic. The pool was located in a dimly lit and soundproof test room with a camera and the experimenter. The maze was divided

into 4 quadrants. Three equally-spaced points around the edge of pool were used as release positions. Order of release positions was varied systematically throughout the experiment. An escape platform (6-cm in diameter and 12-cm high) was located in 1 quadrant, 1 cm above water surface during familiarization session and 1 cm below water surface during other sessions. Video tracking was conducted with a video camera (Sony Dcr-Hc40e) focused on full diameter of pool. The rats were trained in Morris water maze during 5 daily sessions (familiarization session, S1, S2, S3, and S4). The 5 sessions were performed on consecutive days between 9:00 and 12:00. During acquisition phase of experiments, each rat participated in 3 trials per day. For each daily trial, the rat was taken from home cage and placed into the water maze at 1 of 3 randomly

determined locations with its head facing the center of water maze. A trial was started when rat was released from 1 of 3 randomly chosen start positions. After rat found and climbed onto platform, trial was stopped and escape latency was recorded. Maximum trial length was 60 s. If rat had not climbed onto platform within 60 s, the experimenter guided rat by hand to platform and an escape latency of 60 s was recorded. Inter- trial time was 60 s. During this time, rat was kept on escape platform before starting next trial. Rat was then placed in pool again, but at a different location, and next trial began upon its release. Normally, escape latency declines during acquisition as animal learns the location of hidden platform. At the end of third trial, rat was returned to its cage. Twenty-four hours after last acquisition session, a ‘probe trial’ was used

to assess rats’ spatial retention of location of hidden platform. During this trial, platform was removed from the maze, and each rat was allowed to search pool for 60 s before being removed. During this trial, animals should spend more time swimming in quadrant that previously contained the hidden platform than in other 3 quadrants.

After these tests, according to the guidelines of animal ethics committees, the hippocampus and the cortex of brain tissues was excised after the rats were sacrificed. The effect of senescence related inflammation on the tissues in aging rats was evaluated by PCR gene expression analysis. Histological study was performed for cortex and hippocampus tissues.

Statistics
All results are expressed as the mean ±S.E. Acquisition (1–4 days) latency scores in MWM test were measured by 2-way analysis of variance (ANOVA), following post hoc Bonferroni test. Scores of time spent in escape platform’s quadrant in MWM test, first day and retention latencies in PA test, total locomotor activity, and foot shock sensitivity scores were measured by 1-way ANOVA. Criteria for statistical significance was P<0.05. In PCR gene expression analyses, One way ANOVA, Post hoc / Tukey HSD were used and Kruskal Wallis analysis of variance was performed

for data that did not fit the normal distribution.

RESULTS: In MWM test, there’s a significant increase in acquisition latency (1-4 days) of 24-month-old rats. In acquisition session of MWM test, in day 1, there’s a significant difference between aged vs young rats. In day 2, there’s a significant difference between aged vs young rats and aged vs (young +met) group. In day 3, there’s a significant difference between aged vs young rats; aged vs (young +met) group and also (aged+met) group vs aged rats. In day 4, there’s a significant difference between (1) aged vs young rats and (2) aged vs (young+Met) and (3) (aged +Met) vs aged rats.

In probe trial of MWM test, there’s a significant difference between young vs aged rats. There’s a significant difference between (young+Met) vs aged rats and there’s a significant difference between aged vs (aged+met) group. In probe trial of MWM test, there’s a significant reduction in “time spent in the escape platform’s quadrant” in aged rats compared to young rats. Metformin treatment reversed reduction of “time spent in escape platform’s quadrant” of aged rats.

In PA test, there was no significant difference in 1st-day latency of rats in all groups. There’s a significant difference between young vs aged rats and there’s a significant difference between aged vs (aged+Met) group. The locomotor activity of rats was not affected.

A regular cerebral cortex was seen with neurons of normal morphology in the young control and young metformin groups. Degenerated neurons, intercellular and perivascular edema were observed in the aged control group. Regenerated neurons were observed with decreased edema in the aged metformin group. Hippocampus (CA2) with regular morphology was observed in the young control and young metformin groups. Degenerated neurons and severe pericellular and perivascular edema were observed in the aged control group, whereas neurons in the aged metformin group showed regeneration close to normal morphology.

By PCR gene expression analyses there was very significant results in all metformin groups p<0,001; Ki67 & NF- B & TNF- ➔ hippocampus and cortex

PDGFR- β & i-NOS➔hippocampus TGF- & IL-6➔cortex

DISCUSSION: This study demonstrates that chronic metformin treatment affects learning and memory performance in different learning and memory tasks in aged animals. Memory function may be defined as the ability to acquire, process, store, and retrieve information. We investigated the effects of long-term administration of metformin in passive avoidance and Morris water maze test, finding that metformin improved spatial and emotional learning and memory in distinct behavioral tasks involving different brain structures including hippocampus and amygdala.

In our study, we used metformin which beyond its impact on glycemic control, it may have pleotropic effects targeting multiple age-related mechanisms. Previous studies have found that metformin decreases inflammatory markers, NF-κB, ROS and mTOR pathways, thus decreasing DNA damage. Our findings suggest that aging itself, negatively affects spatial and emotional memory functions in rats. Metformin administration might have an affirmative effect on age-related learning and memory dysfunctions. In response to age-related increased inflammatory and

oxidative stress, the neuroprotective effects seen in metformin groups correlate with learning and memory functions. The statistically significant decrease in inflammatory cytokine levels in hippocampus and cortex tissues with metformin and the increase in the levels of proliferation markers with cell-protective properties support the neuroprotective effects of metformin, and these results are consistent with histological findings.