The step-down passive avoidance task is used to evaluate the state-dependent learning and memory. The apparatus consisted of a box made of Plexiglas with dimensions of 40×30×30 cm with a floor of steel bars. Each of the steel bars was 4 mm. in diameter with a spacing of 13 mm. A wooden platform with dimensions of 4×4×4 cm at the bottom center of the floor was provided. Electric shocks with a frequency of 1 Hz at 15 V for 15 s using a stimulator connected to the floor bars transmitted a shock to the animals' hands and feet. When the animal was placed on the podium, the natural tendency of the animal was to get down on the floor bars; however, if the animal received a shock its innate desire to get down from on the platform was suppressed, in particular, an inhibitory avoidance learning occurred. The latency time of moving down from the safe podium (stepdown latency) was considered as memory retrieval [23], which involved two stages, namely training and testing. In the training phase, animals were slowly placed on the wooden platform in the middle of the device and the delay in moving down from the platform was recorded by a chronometer. When the mouse stepped down from the platform and placed all its paws on the grid floor, intermittent electric shocks were delivered continuously for 15 s. Before the final shock, the animal was removed [2425]. This training procedure was performed between 9:00 and 15:00 h, and animals with latencies longer than 30 s were omitted from the study. The test phase was conducted for 24 h after the training phase and was similar to the training phase; however, in this phase the animal did not receive any shock treatment. Thus, each animal was slowly placed on the wooden platform again and the delay in coming down from the platform was considered as a memory retrieval. In these experiments, the time limit for retaining the mouse on the podium was of maximum 300 s [23]. The retention test was also conducted between 9:00 and 15:00 h. At the end of each test, the surface of the apparatus was thoroughly cleaned to avoid the presence of olfactory cues [25].

This test was performed using cross-arm maze. The task was to test the navigation memory, which could be weakened by brain damage. This model was used to test the spatial navigation memory considering the spatial orientation and perception as described by Ragozzino et al. [26]. The maze was constructed of wood, painted gray, and comprised a central platform (25 cm diameter), which radiated four symmetrical arms (55 cm long×10 cm wide) with 12 cm walls. Briefly, mice were placed individually in a four-arm cross maze and were allowed to transverse the maze freely for 12 min. The number and arrangement of entries were noted; an alternation was defined as an entry into four different arms on an overlapping quadruple set. Four consecutive arm choices within the total set of arm choices made up a quadruple set. A quadruple set consisting of arm choices B, D, C, A comprised “actual alternation,” whereas the set with B, D, B, A did not (using this procedure, possible alternation sequences were equal to the number of total arm entries minus three). Percent alternation was calculated as follows:

Alternation percentage is an indicator of spatial navigation memory of the experimental mice and the number of total arm entries is the index of the locomotor activity in this maze [22].

The device comprised a round container, 100 cm in diameter and 50 cm in depth. The tank was filled with water (21–26℃) up to a height of 30 cm and the transparent escape platform made of Plexiglas (10 cm in diameter and 29 cm in height) was concealed at 1.5 cm under the surface of water at a fixed place. The water was turned milky with powdered nonfat milk or nontoxic white colored dye. The platform was not observable from just above the water level and removal trials have designated that escape on to the platform was not done by visual or other proximal signs [2728]. The period spent by the animal to reach the concealed platform was used as the memory index. Before the test commencement, the mice were adapted to the maze location. The water maze test was conducted for all mice groups on 10th, 20th, 30th, and 40th days for all animals (n=5). For each trial, the time required (in seconds) for an individual mouse to find the concealed platform was recorded and the mean data from the tests were used for the statistical analyses.

At the end of 40 days, the mice were decapitated under ether anesthesia. The skull was cut open and the brain was exposed from its dorsal side. The whole brain was quickly removed and weighed. Then the ratio of brain weight to body weight was calculated for each animal [29].

The whole brain was quickly removed and cleaned with normal saline on the ice. A 10% (w/v) homogenate of brain samples (0.03 M sodium phosphate buffer, pH-7.4) was prepared by using an Ultra-Turrax T25 (USA) homogenizer at a speed of 9500 rpm. The homogenized tissue preparation was used to measure AChE, ROS, Catalase, MDA, GSH and nitrite.

The level of ROS in brain tissue was measured using 2, 7-dichlorofluorescindiacetate (DCFDA) that converted into highly fluorescent DCF by cellular peroxides. Briefly, 10% brain homogenate were prepared in phosphate buffer 1 mM, pH 7.4. For each test 2 ml homogenate tissue was mixed with 40 ml of 1.25 mM DCFDA in methanol for ROS estimation. All samples were incubated for 15 min in a water bath at 37℃. Fluorescence was calculated using a fluorimeter, at 488 nm excitation and 525 nm emission wavelength [30].

Glutathione content was measured according to the method described by Thomas and Skrinska. Brain homogenates were incubated whit 1 ml of 20% trichloroacetic acid (TCA) and 1 ml EDTA 1mM for 5 min, which was used as protein precipitant. The total homogenate was centrifuged at 10,000 g for 30 min at 4℃. 200 µl of supernatant was mixed with 1.8 ml of 0.1 mM 5.5′-dithio-bis (2-nitro benzoic acid) (DTNB). The GSH reacts with DTNB and forms a yellow-colored complex with DTNB. The absorbance was read at 412 nm. The result was expressed as µmoles of GSH/mg protein [31].

The extent of lipid peroxidation in terms of malondialdehyde (MDA) formation was measured. Briefly homogenate brain sample containing 1 ml was mixed with 1 ml TCA (20%), 2 ml TBA (0.67%) and heated for 1 h in boiling water bath. After cooling, mixture centrifuged and absorbance of the supernatant measured at 532 nm against suitable blank. The amount of TBARS was calculated by using a molar extinction coefficient of ε=1.56×10⁵/M/cm and expressed as mol/mg protein [31].

Catalase activity was assayed according to the method used by L.Goth. 500 µl of 0.05 mmol Tris-HCl, 1 ml H₂O₂ and 50 µl of sample were mixed and incubated for 10 min, and then Reaction was stopped by adding 500 µl Ammonium molybdate solution 4%. The absorbance was read at 410 nm. The result was expressed as U/mg protein [32].

The SOD activity was determined using a xanthine/xanthine oxidase system for production of superoxide radical and subsequent measurement of cytochrome c as a scavenger of the radicals. Optical density was evaluated using a spectrometer (UV-1601, Shimadzu) at 550 nm. One unit of enzyme activity was defined as the quantity of SOD required to inhibit the reduction rate of cytochrome c by 50% [33]. SOD activity is presented as units per milligram of protein (U/mg protein).

Total brain protein was estimated by the Lowry et al method using bovine serum albumin (BSA) (1 mg/ml) as standard. The absorption was read spectrophotometrically at 750 nm [34].

Nitrite was estimated using Greiss reagent, which served as an indicator of nitric oxide production. An amount of 100 µL Greiss reagent (1:1 solution of 1% sulphanilamide in 5% phosphoric acid and 0.1% napthaylamine diamine dihydrochloric acid in water) was added to 100 µL of supernatant and absorbance was measured at 542 nm [35]. Nitrite concentration was calculated using a standard curve for sodium nitrite. Nitrite levels were expressed as of the control percentage.

The whole brain AChE activity was measured by the Ellman method [36] and Voss and Sachsse method [37]. Change in the absorbance per min of the sample was read spectrophotometrically at 420 nm.

Twenty four hours after the last experimental day, the overnight fasting animals were anesthetized by ether. Fasting blood glucose levels (FBG) were measured using a glucometer (Elegance CT-X10, Convergent Technologies, Germany) by cutting the tail tip of mice. Furthermore, blood samples were directly collected by cardiac puncture and centrifuged at 3500 rpm for 20 min. Insulin level was measured by ELISA assay kits (Monobind, USA) (The sensitivity of hormone detection per assay tube was 0.182 µIU/ml).

Increased ROS formation is expressed as DCF fluorescence intensity unit. As mentioned in Table 1, ROS level was significantly (p<0.001) elevated in the brain of diabetic animals as compared to the control. Cinnamic acid administration significantly (p<0.001) inhibited the ROS production in diabetic mice. Moreover, exposure to cinnamic acid resulted in significantly (p<0.05) lower ROS formation in nondiabetic mice when compared with the control (Table 1).

Glutathione assessment results reported a significant decrease in the diabetic group compared to the control mice (p<0.001). Cinnamic acid administration significantly (p<0.01) prevented this decrease in GSH level in diabetic mice. Moreover, exposure to cinnamic acid resulted in significantly (p<0.05) greater GSH level in nondiabetic mice when compared with the control mice (p<0.05) (Table 1).

The results of lipid peroxidation revealed that STZ treated group reported significant increase in the MDA level (p<0.001) in comparison to the control group. Cinnamic acid administration significantly (p<0.01) inhibited the MDA level in diabetic mice. No significant difference was observed between cinnamic acid-treated mice and control group (Table 1).

Assessment of brain homogenates indicated a significant (p<0.01) decrease in the CAT activities in diabetic group than the control animals, while cinnamic acid administration significantly (p<0.01) ameliorated these indices in diabetic mice. No significant difference was observed between the cinnamic acid-treated mice and the control group (Table 1).

As shown in Table 1, the SOD production was significantly reduced in brains of the STZ treated mice (p<0.001). Administration of cinnamic acid significantly increased the SOD activity in the brain, when compared with the STZ-treated group (p<0.01).