A New Vanadium Complex Improves the Spatial Learning and Memory by Activation of Caveolin – MAPK – CREB Pathway in Diabetic Mice

Diabetes mellitus, a common metabolic disease with a rising global prevalence, is associated with long-term complications of peripheral nervous system and the central nervous system [1-6]. Numerous studies indicate that diabetes mellitus might be accompanied with a certain damage of brain function involved in impairment of attention, information processing speed and spatial learning and memory [7-11]. The pathogenesis of the impaired brain function relating to diabetes has been studied involved in dysregulated innate immunity [12], vascular and metabolic mechanism [13], hippocampal neuronal apoptosis [14] and other factors. However to date there is still no convincing empirical evidence on the pathogenesis of this disorder and the drug candidates for improving cognitive deficits caused by diabetes are still few.


Introduction
Diabetes mellitus, a common metabolic disease with a rising global prevalence, is associated with long-term complications of peripheral nervous system and the central nervous system [1][2][3][4][5][6]. Numerous studies indicate that diabetes mellitus might be accompanied with a certain damage of brain function involved in impairment of attention, information processing speed and spatial learning and memory [7][8][9][10][11]. The pathogenesis of the impaired brain function relating to diabetes has been studied involved in dysregulated innate immunity [12], vascular and metabolic mechanism [13], hippocampal neuronal apoptosis [14] and other factors. However to date there is still no convincing empirical evidence on the pathogenesis of this disorder and the drug candidates for improving cognitive deficits caused by diabetes are still few.
The transcription factor cAMP responsive element binding protein (CREB) is a nuclear protein that modulates the transcription of genes with cAMP responsive elements in their promoters. Genetic and pharmacological studies in mice and rats have demonstrated that CREB is essential for the induction of many forms of long-term synaptic plasticity, including spatial and social learning [15,16]. CREB1 is a transcriptional activator necessary for induction of long-term synaptic facilitation, while CREB2 functions as a transcriptional repressor that poses inhibitory constraints on the induction and formation of longterm memory. The repression by CREB2 could be relieved, possibly via phosphorylation of CREB2 by mitogen-activated protein kinase (MAPK) [17]. CREB1 and CREB2 interact via interlocked positive and negative loops [18]. Since CREB is critical for converting short-to longterm memory, the proteins in CREB pathway have been demonstrated as important targets for memory-modifying drugs [19].
Insulin/insulin receptor (IR) plays diverse roles in brain functions via activating specific signal transduction cascades, including spatial learning and memory formation [20,21]. As a receptor tyrosine kinase, IR in the hippocampus has been shown responding to learning experiences by alterations in its gene expression and activation of downstream molecules such as Shc/Erk1/2 and IRS-1/AKT in the stage of memory formation [20,21]. Caveolae, a subset of membrane microdomains, and its resident protein Caveolin-1 (Cav-1) are reported playing a major role in insulin signaling. IR is highly concentrated in caveolae and it has been reported that there is a critical important interaction between Cav-1 and IR in executing successful insulin signaling in adipocytes [22]. Recently, Cav-1 has been demonstrated to be essential for estrogen receptor α activation of metabotropic  [23]. In addition, our previous results have shown that Cav-1 was involved in the discrimination learning [24].
Vanadium, a required trace element of human body, has been proved to have insulin-like function [25]. Since the insulin-like effects of vanadium salts in isolated rat adipocytes were reported, the insulinlike actions of vanadium complex have been examined in a large variety of insulin-responsive cells and tissues [26]. Vanadium complex not only improved the hyperglycaemia but also were involved in the treatment of diabetic complication such as obesity and hypertension [26][27][28]. Recently, more and more studies focused on the role of vanadium complex in brain function. It was reported that vanadium complex protected the streptozotocin-induced oxidative damage [29] and ameliorated the altered antioxidant status and membrane linked functions in diabetic rat brains [30]. Recently, vanadium-enriched chickpea was reported to ameliorate some hyperglycemic symptoms of the diabetic rats and reduce diabetes relating spatial learning and memory impairment [31]. Study also indicated that vanadium complex could be viewed as potential therapeutic agent to enhance ischemiainduced neurogenesis through PI3K/AKT and Erk activation [32]. However, potential short and long-term vanadium toxicity has slowed the acceptance for therapeutic use [33].
With all above background, we hypothesize that the new vanadium complex may ameliorate the learning and memory of diabetic mice and its mechanism may be involved in Cav-1-PI3K/AKT or Erk-CREB activation. In present study, to investigate the effect of the new vanadium on diabetes relating spatial learning and memory impairment, the vanadium complex were administrated to diabetic mice, which were induced by alloxan. The expression of proteins related to learning and memory in hippocampus of mice was examined by Western blot following Morris water maze.

Animals
All animal experiments followed the guidelines of the International Council for Laboratory Animal Science (ICLAS). Efforts were made to minimize animal suffering and only the number of animals necessary to produce reliable scientific data was used. 3-5 months male Kunming mice (weighing 20±2 g, specific pathogen free) were purchased from Dalian Medicial University (Dalian, China). The mice were housed five per cage at 23±2°C (12-h light/dark cycle) with ad libitum access to food and water. Mice were sacrificed by decapitation immediately after behavioral test. Hippocampus tissues were isolated and stored at -80°C until processed for further processing.
Horseradish peroxidase-labeled goat anti-rabbit and horseradish peroxidase-labeled goat anti-mouse secondary antibodies were from Boster Biotechnology (China). Other biochemical reagents were from Sigma (USA) and Promega (USA).
Cells were pre-incubated in 96 multi-well plates at a density of 5×10 3 cells/well. When cells reached 70 % confluence, the monolayers were washed twice with DMEM. Cells were incubated in the medium with vanadium complexes of 10 μМ for 24 h. Then the culture medium was replaced by a solution of MTT (5 mg/mL) in serum-free growth medium. After 4 h incubated at 37°C in a 5 % CO 2 atmosphere, the supernatant was poured off, and the reaction was stopped with DMSO. The extracted was quantified at 570 nm with a microplate reader (Multiskan Ascent, Thermo Labsystems, USA), and the percentage viability was calculated.

Diabetes model induction and administration of vanadium complex
A single dose of 200 mg/kg alloxan freshly prepared in 20 mL/kg 0.9 % saline buffer was injected intraperitoneally to induce diabetes. The control mice received the same volume of 0.9 % saline buffer. Diabetes was confirmed after 3 days of alloxan injection for four times. After 48 h following alloxan injections, the blood samples were collected through tail vein and blood glucose levels were estimated by Glucoval Compact Meter and Glucoval Strips (Biochemical Systems International s. r. l., Italy). On day 7 following alloxan injections, the mice with fasting blood glucose levels 540 mg/dL or higher were considered diabetic.
Diabetic mice were randomly divided into diabetes group (Diabetes, n=8) and vanadium treatment group (Vanadium, n=7). Vanadium group were treated with vanadium complex (5 mg/kg, p.o.) for consecutive three weeks. Vanadium complex was dissolved in 0.9 % saline buffer before administration in a constant volume of 15.6 mL/ kg body weight. Control group (Control, n=10) and diabetes group received 15.6 mL/kg 0.9 % saline buffer. Glucose level was estimated weekly. Body weight was measured daily.

Open-field test
The mice were evaluated in big square box (100 cm×100 cm×50 cm) for a period of 5 min. The bottom of testing box was divided into 25 grids. The testing box was wiped clean with ethanol after each test. Following each mouse adapting to the box environment for 1 min, spontaneous locomotor activity was evaluated. We chose the number of crossing grids and the rearing times within 5 min as indicators [24].

Morris water maze test
After open-field test, animals were tested in a spatial version of Morris water maze test [35]. The apparatus consisted of a circular water tank (100 cm in diameter and 50 cm high). A platform (9 cm in diameter and 29 cm high), invisible to the mice, was set inside the tank which was filled with water maintained at approximately 20±1°C at a height of 30 cm. The tank was located in a large room where there were several different colored cues external to the maze; these were visible from the pool and could be used by the mice for spatial orientation. The position of the cues remained unchanged throughout the study.
The water maze task was carried out for 5 consecutive days. The mice received four daily training trials, with each trial having a ceiling time of 60 s and a trial interval of approximately 20 min. For each trail, each mouse was put into the water at one of four starting positions, the sequence of which being selected randomly. During test trials, mice were placed into the tank at the same starting point, with their heads facing the wall. The mouse had to swim until it climbed onto the platform and remained there for 10 s. The escape platform was kept in the same position relative to the distal cues. If the mouse failed to reach the escape platform within the maximally allowed time of 60 s, it was gently placed on the platform and allowed to remain there for the same amount of time.
On the next day a probe trial was performed where in the extent of memory consolidation was assessed. The time spent in target quadrant indicates the degree of memory consolidation that has taken place after learning. In probe trial, the mice were placed into the pool as in the training trial, except that the hidden platform was removed from the pool.
The protein content of the samples was measured using the Coomassie technique. 100 μg of total protein were separated by 12 % sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) 2 h and transferred onto PVDF membrane (Hybond, USA) at room temperature (220 V for 40 min). Equal Quantitative measurement of protein level was performed using the Image J program (NIH). And the value of protein levels was designed as 1 in the control group. The results were expressed as mean proportion of the control group values. Experimental design was shown in ( Figure  1).

Statistical analysis
One-way ANOVA followed by Tukey test for multiple pairwise examinations was used for comparison. Changes were identified as significant if p was less than 0.05. p<0.01 was taken to indicate marked difference. Mean values were reported together with the standard difference of the mean (S.D.).

Lower-toxicity of the new vanadium complex with 3,5-dimethyl-pyrazolyl ligand
To evaluate the toxicity of the new oxovanadium complex with 3,5-dimethyl-pyrazolyl ligand compared with other oxovanadium complexes, MTT reduction assay was conducted on 3T3-L1 cells.
As shown in (Figure 2), incubation with 10 μM VO(ph-acac) (HB(pz) 3  Open-field test Sacrifice Figure 1: Experimental design. Diabetic mice were induced by four times injections of alloxan. One week after alloxan injections, mice with blood glucose over 540 mg/dL were considered diabetes. Mice were randomly divided by three groups, which are control group, diabetes group and vanadium group (Diabetes+Vanadium). Vanadium group were treated with vanadium complex daily for three weeks. And then open-field test and Morris Water Maze test were performed. After behavioral experiment, animals were sacrificed by decapitation. symptoms of diabetes. There was a marked decline in the body weight and a dramatic increase in blood glucose level in diabetic mice compared with control mice (p<0.01). Treatment of diabetic mice with the new vanadium complex effectively increase the body weight to normal level and lowered the high blood glucose level (p<0.05) ( Table 1).

Improved performance of diabetic mice in Morris water maze by vanadium complex
Before testing cognitive function of the three groups, spontaneous locomotor activity was tested and the results showed no difference in the three groups' mice ( Table 2).
The learning and memory ability was assessed by Morris water maze. The mean escape latency of all the trained mice decreased over the course of the five days learning trials. There is no difference between diabetes and control group on the first and second days. But the mean escape latency of diabetic mice was longer than that of control mice in the third and fifth days. Vanadium treatment shortened the mean latency in diabetic mice ( Figure 3). Diabetic mice showed a lower ability to find the platform and learn its location. This poorer performance was improved by the vanadium treatment. Although there is no marked difference in the escape latency, the percent time in target quadrant of the three groups suggests the same performance. The percent time in target quadrant did not differ between any of the groups on first and second days. But from the third day onwards there was significant Blood glucose level of control, diabetic and vanadium-treatment mice were measured on a weekly basis and the body weight of each group were measured daily. The final measured results at the end of three weeks vanadium treatment are shown. Data are presented as mean±S.D. **p<0.01 vs. Control, p<0.05 vs. Diabetes.    difference in the three groups. Compared with that in control group, the percent time in target quadrant was lower (p<0.05) in diabetes group. After vanadium treatment the percent time in target quadrant was increased (p<0.05) and equivalent with that in control group (Figure 4).
In the probe trial of the Morris water maze study, which measures how well the mice had learned and consolidated the platform location during the five days' training, differences were shown in the three groups ( Figure 5). Control mice spent almost half of the time in the target quadrant while diabetic mice spent equal time in each quadrant. Subtle increase of percent time spent in target quadrant was shown in vanadium treatment mice compared diabetic mice, but it is obvious that the orientation of platform in vanadium treatment mice was not exact. The results showed that diabetic mice were accompanied with impaired spatial memory retention and vanadium had subtle improvement on the impaired spatial memory retention.

Changes of protein expressions by the new vanadium complex in diabetic hippocampus
To investigate the mechanism of improvement of vanadium on the impaired learning and memory ability in diabetes, here we examined the expression of proteins involved in insulin signaling by Western blot. The expressions of AKT and phospho-AKT did not significantly differ in each group (Figure 6), while the phosphorylation level of p42/p44 MAPK protein was remarkably decreased in diabetes group and then increased after vanadium treatment (Figure 7). The results suggest that MAPK signal pathway might be involved in the effect of vanadium on the learning and memory in diabetic hippocampus. In addition, the expression of upstream protein Cav-1 and downstream transcription factor CREB2 were following examined. Cav-1 expression was remarkably reduced and while after vanadium treatment Cav-1 expression was significantly increased. The expression of CREB2, which inhibits long-term memory, was higher in diabetic and vanadium treatment mice and vanadium failed to change the CREB2 expression in diabetes (Figure 7).

Discussion
Large amounts of studies suggest that diabetes mellitus are accompanied with impaired cognitive functions [7][8][9][10][11]. The potential mechanisms for this include direct effects of hypo-or hyperglycaemia, but also indirect effects via cerebrovascular alterations [36,37]. Vanadium has been proved to have insulin-like function. It could improve the hyperglycaemia, protect the streptozotocin-induced oxidative damage [29] and ameliorate the altered antioxidant status and membrane linked functions in diabetic rat brains [30]. This study analyzed the role of vanadium on the learning and memory of diabetic mice. Alloxan-induced diabetes produced marked impairment in learning and memory ability. Chronic treatment with vanadium significantly ameliorated the declined learning ability of diabetes in the acquisition experiment. But vanadium had subtle improvement on the impaired spatial memory retention of diabetes in the probe trial. Vanadium could improve the hyperglycaemia but also its anion could penetrate blood-brain barrier [38]. Thus, we couldn't ensure which of the antidiabetic and direct effect of vanadium or both contributed the improvement on diabetic learning and memory. In present study, the diabetic mice only obtained three weeks treatment by vanadium and their glucose levels remain over 540 mg/dL though have been significantly declined by vanadium. On the other hand, vanadium had neurotoxic effects on the brain in some extent, though the vanadium complex with lower toxicity was used in the experiment [38]. The two aspects may contribute to the subtle improvement on the long-term memory.
Evidence has shown that the insulin and insulin receptor (IR) play a role in cognitive function. IR signaling may play a modulating role to aid functions of the mainstream neurotransmitter receptors during memory processing. Learning-specific increases in levels of downstream molecules such as IRS-1 and AKT were detected in the synaptic membrane accompanied by decreases in AKT phosphorylation. Translocation of Shc protein to the synaptic membrane and activation of Erk1/2 were also observed after long-term memory formation [21]. In our results, p42/p44 MAPK phosphorylation was decreased in diabetic mice which were deficient in learning ability and spatial memory retention while after vanadium treatment it was remarkably increased. The expressions of AKT and phospho-AKT did not differ in each group. The results suggest that MAPK signal pathway may involve in the improvement of vanadium on the learning and memory ability in diabetic hippocampus.
Activation of Erk1/2 was observed after long-term memory formation [21]. In our study, vanadium increased p42/p44 MAPK phosphorylation in diabetic mice but the impaired spatial memory retention of diabetes was subtly improved by vanadium. Therefore, we inferred that the downstream cascades of MAPK involved in long-term memory may be not changed by vanadium. The protein expression of CREB2, a long-term memory inhibitory transcription factor [39], was detected in the hippocampus of each group by Western blot. As inferred, the expression of CREB2 was increased in diabetic mice compared with that in control mice, and was not significantly changed by vanadium treatment. It convinced the observation that impaired spatial memory retention of diabetes was subtly improved by vanadium.
Cav-1, a protein marker of caveolae, was reported playing a major role in insulin signaling and involved in discrimination learning [22][23][24]. Cav-1 expression was also examined in present study. The decline of Cav-1 expression was shown in diabetic mice and after vanadium treatment Cav-1 expression was significantly increased. IR was resident in caveolae via its binding to the scaffolding domain of Cav-1 through the caveolin binding domain in its cytoplasmic region. During the state of insulin resistance the IR and Cav-1 complex was dissociated [22]. In diabetes, IR and Cav-1 complex may be dissociated and MAPK signaling pathway is disrupted with deficient p42/p44 MAPK phosphorylation. As a result, nuclear transcription factor CREB2 is activated as a longterm memory inhibitor. Vanadium repaired the declined learning and memory via this signaling pathway.
In conclusion, this new vanadium complex treatment improved the learning and memory ability in these diabetic mice. The mechanism may be involved in the activation of Caveolin-MAPK-CREB pathway in the neuron. Moreover, due to its relative lower toxicity and high hypoglycemic effect efficiency, organic vanadium may find clinical application in treating neuronal disturbances in the diabetic patients. More studies are also needed to establish the efficacy and safety of the complex prior to its use in humans.