Insulin Receptor Function In Rat Hippocampus Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Cardiac Electrophysiology Research and Training Center, Department of Physiology, Faculty of Medicine, †Department of Oral Biology and Diagnostic Science, Faculty of Dentistry, Chiang Mai University, Chiang Mai 50200 Thailand

The development of abdominal obesity following chronic consumption of a high-fat diet contributes to peripheral insulin resistance. Although the relationship between peripheral insulin resistance and cognitive impairment have been shown, the effect of high-fat consumption and the neurofunctional insulin sensitivity in CA1 hippocampus is unclear. We tested the hypothesis that high-fat diet consumption can lead to peripheral insulin resistance and impair neuronal responses to insulin by using extracellular recording in CA1 hippocampus and the immunoblot technique to determine the neuronal function of insulin receptors (IRs) in rats. In 12- week high-fat-fed (HF) rats, peripheral insulin resistance was observed, but was not found in 4-or 8-week HF rats. The neuronal IR response demonstrated by insulin-mediated long-term depression (LTD) in CA1 hippocampus was diminished in 12-week HF rats. This reduction of insulin-mediated LTD correlated with various parameters of peripheral insulin resistance (p<0.05). However, the impairment of insulin-mediated LTD in hippocampus did not cause any change in paired-pulse ratio and carbachol-induced LTD between both dietary groups, suggesting that the defect of neuronal insulin receptors had no effect on the presynaptic transmission and did not interrupt other forms of synaptic plasticity. Furthermore, levels of phosphorylation of neuronal IR, neuronal IR substrate 1 (IRS-1) and neuronal Akt/PKB in response to insulin were significantly decreased in 12-week HF rats without any change in the level of: IR, IRS-1 and Akt/PKB protein. These findings suggest that neurofunctional insulin resistance can develop at the same time as peripheral insulin resistance in HF rats. The neuronal insulin resistance may lead to neuronal ageing as shown in the reduction of nNOS-immunoreactive neurons in 12-week HF rats with the impairment of neuronal insulin receptors.


A high-fat diet has been shown as the major cause of obesity and insulin resistance (1). Surprisingly, considering the dramatic rise of insulin resistance throughout the developed nations (2, 3) and the growing interest in the role of insulin within the brain (4-7), there have been only a few studies examining the effects of this metabolic impairment in the central nervous system (CNS). Growing evidence has shown that insulin resistance, defined as a sharply diminished insulin receptor (IR) response to insulin within target tissues, has grown increasingly common in obese people (8, 9).

The expression of insulin receptors (IRs) is found throughout the body, in organs or cells such as liver, muscle, fat, red blood cells and neurons in the CNS (10-12). In the CNS, insulin has been shown to regulate neurotransmitter release and synaptic plasticity (13, 14), and the impairment of insulin signaling in the CNS has been shown to relate to neurodegenerative diseases (6, 15-19). Evidence from clinical and animal studies suggests that insulin/IR signaling may play a role in learning and memory. Zhao and colleagues found that the up-regulation of IR in CA1 hippocampus is associated with short-term memory formation after a spatial learning experience (20). It has also been shown that insulin is required to produce memory improvement in elderly people and patients with Alzheimer's disease (21, 22). Furthermore, clinical studies have shown that cognitive impairments are often found in association with increased insulin resistance (23-25).

The underlying mechanisms of IRs in learning and memory could involve the relationship between IRs and the synaptic plasticity in the CNS. Insulin has been shown to play a role in synaptic plasticity by acting on alpha-amino-3 hydroxy-5-methylisoxazole-4-propionic acid (AMPA) receptor trafficking (26-29). It has been demonstrated that insulin facilitates clathrin-dependent internalization of AMPA receptors, causing long-term depression (LTD) of AMPA receptor-mediated synaptic transmission in hippocampal CA1 neurons (26, 28). This piece of evidence indicates that the neurofunctional IRs in the CA1 hippocampus is insulin-mediated LTD. Moreover, the activation of insulin turns on the protein kinase activity of the IR, which triggers cascades of signal transduction through its downstream substrate molecules. Several insulin receptor signaling pathways activated by IRs include insulin receptor substrate-1 (IRS-1) (30). It has been demonstrated that rats trained in a spatial learning task showed the learning-specific increase in IRS-1 in the hippocampal synaptic membranes (31). These findings suggest that IR signaling plays a role in learning and memory by modulating activities of synaptic plasticity such as insulin-mediated LTD and by triggering signal transduction cascades such as IR, IRS-1 and Akt/PKB.

Diets have been shown to influence cognitive functions. In insulin resistance caused by fructose-fed hamsters, it has been shown that hippocampal synaptic plasticity, an important biological mechanism of learning and memory, was impaired (32). Excessive fat consumption has also been shown to play important and integral roles in the development of insulin resistance and type 2 diabetes (33). A recent study demonstrated that consumption of a high-calorie diet for 32 weeks reduced hippocampal synaptic plasticity and impaired cognitive function in rats (34). Furthermore, several studies suggest that consumption of a diet rich in fat for 3 months can develop peripheral insulin resistance and impede cognitive performance (6, 7, 35-37). These findings suggest that the development of insulin resistance can mediate the cognitive deficit associated with high-fat diet. However, the effect of high-fat diet on the neurofunctional IRs is still unclear. In addition, the effects of time-course of high-fat diet consumption on the neurofunctional IRs have never been investigated. Therefore, in this study we tested the hypothesis that high-fat diet consumption for a specific period of time can cause peripheral insulin resistance and can lead to impaired neuronal response to insulin (or neuronal insulin resistance). We used an electrophysiological study to investigate whether the neuronal responses to insulin (insulin-mediated LTD) are altered by high-fat diet consumption at different time course in order to detect the earliest stage of the disruption of the neurofunctional IRs. We also examined the alteration of biochemical activity of insulin receptor pathways: IRβ, IRS-1 and Akt/PKB, in the brain following each time course of high-fat diet consumption. Furthermore, we investigated whether the neuronal insulin resistance leads to neuronal ageing using the amount of nNOS-immunoreactive neurons as an indicator of neuronal ageing.

Materials and methods

Animals and dietary protocols

All experiments were conducted in accordance with an approved protocol from the Faculty of Medicine, Chiang Mai University Institutional Animal Care and Use Committee, in compliance with NIH guidelines. Male Wistar rats weighing ~ 180-200 g were obtained from the National Animal Center, Salaya Campus, Mahidol University, Thailand. All animals were individually housed in a temperature-controlled environment with a 12:12 light-dark cycle. One week after arrival, rats were randomly assigned to one of the two dietary groups (n=43 in high-fat diet group and n=44 in normal diet group). The normal-diet (ND) group received a standard laboratory chow, in which 19.7% of total energy (%E) was from fat, with energy content calculated at 4.02 kcal/g (Mouse Feed Food No. 082, C.P. Company, Bangkok, Thailand). The high-fat (HF) group consumed a high-fat diet, containing fat, mostly from lard (59.3% E), with energy content calculated at 5.35 kcal/g, for 12 weeks. The animals were maintained in individual cages with unrestricted access to food and water. Body weight and food intake were recorded daily. Blood samples were collected from the tail at weeks 4, 8 and 12 after fasting for at least 5 hours. Samples for glucose assay were kept on ice in tubes precoated with sodium fluoride. Samples for insulin and triglyceride assay were taken in tubes with EDTA. Plasma was separated and stored at -80oC for subsequent biochemical analyses. At the end of each experimental period (4, 8 and 12 weeks) of both dietary regimens, animals were deeply anesthetized and decapitated. The brain was rapidly removed for brain slice preparation and one lobe of liver as well as visceral fat were removed, weighed and stored at -80oC for further biochemical analysis.

The oral glucose tolerance test (OGTT) was investigated in the 12-week high-fat diet and 12-week normal diet group. After rats being on the dietary for 12 weeks, animals were fast for 12 hours before they were used in the OGTT. An OGTT consisted of 2 g/kg body weight glucose feeding by gavage. Blood (0.25 ml) was collected from a small cut at the tip of the tail immediately before and at 15, 30, 60 and 120 min after glucose feeding. Whole blood was mixed with EDTA and centrifuged at 10,000 rpm to isolate the plasma. The plasma was stored at -80oC until it was used for glucose analysis with a commercially available kit (Biotech, Bangkok, Thailand).

Analytical Procedure

Fasting plasma glucose and triglyceride concentrations were determined by colorimetric assay using commercially available kits (Biotech, Bangkok, Thailand). Fasting plasma insulin level was measured by Sandwich ELISA kits (LINCO Research, Missouri, USA).

Determination of insulin resistance (HOMA index)

Insulin resistance was assessed by Homeostasis Model Assessment (HOMA) (38, 39) as a mathematical model describing the degree of insulin resistance, calculated from fasting plasma insulin and fasting plasma glucose concentration. A higher HOMA index indicates a higher degree of insulin resistance. The HOMA index was determined by the following equation:

[fasting plasma insulin (µU/ml)] x [fasting plasma glucose (mmol/l)]


Analysis of liver triglyceride concentration

Tissue homogenates were prepared for triglyceride assay by a modification of the method of Frayn and Maycock (40). A 100-200 mg portion of liver was minced and put into a glass tube containing 3 ml of chloroform-isopropanol 7:11 (v/v). The homogenate was left at room temperature for at least 16 hours. Then, 1 ml of homogenate was pipetted into a glass tube and evaporated to dryness at 40oC for 16 hours. The dried residue was dissolved and mixed in 10% bovine serum albumin. The triglyceride concentration was analyzed with a commercially available kit (Biotech, Bangkok, Thailand).

Brain slice preparation

At the end of weeks 4, 8 and 12, the animals were anesthetized with isoflurance after fasting for at least 5 hours. After being decapitated, the brain was removed and immersed in ice-cold "high sucrose" aCSF containing (mM): NaCl 85; KCl 2.5; MgSO4 4; CaCl2 0.5; NaH2PO4 1.25; NaHCO3 25; glucose 25; sucrose 75; kynurenic acid 2; ascorbate 0.5, saturated with 95%O2/5%CO2 (pH 7.4). This solution enhanced neuronal survival during the slicing procedure (41). Hippocampal slices were cut using a vibratome (Vibratome Company, St. Louis, Mo., USA). Following a 30-minute post-slice incubation in high sucrose aCSF, slices were transferred to a standard aCSF solution containing (mM): NaCl 119; KCl 2.5; CaCl2 2.5; MgSO4 1.3; NaH2PO4 1; NaHCO3 26; and glucose 10, saturated with 95% O2/5%CO2 (pH 7.4) for an additional 30 minutes, before being used for the extracellular recordings, immunoprecipitation and immunoblotting.

Extracellular recording of hippocampal slices

To investigate insulin-induced long-term depression (LTD), the hippocampal slices were transferred to a submersion recording chamber and continuously perfused at 3-4 ml/min with standard aCSF warmed to 25-28°C. Field excitatory postsynaptic potentials (fEPSPs) were evoked by stimulating the Schaffer collateral-commissural pathway with a bipolar tungsten electrode, while recordings were gathered from the stratum radiatum of the hippocampal CA1 region with micropipettes (3 Mohm) filled with 2M NaCl. Stimulus frequency was 0.033 Hz. The stimulus intensity was adjusted to yield a fEPSP of 0.8-1.0 mV in amplitude. Hippocampal slices were perfused with aCSF (as baseline condition) for 10 minutes and then perfused with aCSF plus 500 nM insulin (as insulin stimulation) for 10 minutes, after which the hippocampal slices were perfused with aCSF again (wash out) and recorded for the next 30 minutes.

To investigate that the reduction of insulin-mediated LTD was the result of an alteration of neuronal insulin signaling and not a non-specific alteration of synaptic transmission following neuronal insulin resistance, we examined the characterized form of synaptic plasticity that may not depend upon insulin signaling, by measuring paired pulse facilitation (PPF). PPF is a measurement of short-term potentiation, which may occur at the presynaptic sites (42). PPF occurs following two identical stimulations, separated by 50-msec, applied to the Schaffer collateral. PPF was shown as the increased response observed in the second stimulation, compared to the first one.

To investigate carbachol-induced LTD (mLTD), the protocol for inducing mLTD was as following. The stable 10-min baseline of CA1 extracellular field excitatory postsynaptic potentials (fEPSPs) were recorded by stimulating the Schaffer collateral-commisural pathway with bipolar tungsten electrode. Stimulus frequency was 0.1 Hz and stimulus intensity was adjusted to yield fEPSPs of 0.8-1.0 mV amplitude. The cholinergic agonist, carbachol 50 µM (Calbiochem, San Diego, CA, USA), was superfused for 10 minutes to induce mLTD after which the carbachol was washed out and recorded for the next 30 minutes.

All data were filtered at 3 kHz, digitized at 10 kHz, and stored on a computer using pClamp 9.2 software (Axon Instruments, Foster City, CA, USA). The initial slope of the fEPSPs was measured and plotted versus time using Origin 8.0 software.

Preparation of brain homogenates for immunoprecipitation and immunoblotting

To examine the alteration of neuronal insulin-mediated phosphorylation of the IR, the IRS-1 and the Akt/PKB following 4, 8 and 12 weeks of two dietary regimens, six brain slices per animal were placed into either aCSF or aCSF plus 500 nM insulin (Humelin R, Eli Lilly, Giessen, Germany) for 5 minutes. Then, three brain slices in each conditioned group were homogenized in 500 μl of ice-cold brain slice lysis buffer [1mM EDTA, 1mM EGTA, 1% NP-40, 1% Triton X-100 and supplemented with a protease inhibitor cocktail, Roche complete mini-tablets, (Roche Molecular Biochemicals, Indianapolis, IN, USA)]. Next, the homogenates were centrifuged at 9,000 g for 30 minutes at 4oC and the protein concentration was measured using the Bio-Rad DC Protein assay kit (Bio-Rad Laboratories, Hercules, CA, USA). These homogenates were then stored at -80oC for further biochemical analysis of the tyrosine phosphorylation of IR, IRS-1 and Akt/PKB.

To determine the level of IR, IRS-1 and Akt /PKB protein expression in the brain, another set of three brain slices in aCSF were homogenized over ice in non-ionizing lysis buffer containing: 100mM NaCl, 25mM EDTA, 10mM Tris, 1% Triton X-100, 1% NP-40 supplemented with a protease inhibitor cocktail (Roche Molecular Biochemicals). Then, homogenates were stored at -80oC for further biochemical analysis of IR and IRS-1.

Immunoprecipitation and Immunoblotting

IRS-1 protein and tyrosine phosphorylation of IR and IRS-1 were immunoprecipitated from brain homogenates with polyclonal antibodies against each protein (1 μg antibody/ 500 μg total lysate). Rabbit anti-IR and rabbit anti-IRS-1 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) with protein A agarose beads were used to prepare each protein for immunoprecipitation as previously described (32). After an overnight incubation at 4oC, all samples were centrifuged and the supernatant was removed. The beads were washed three times with ice-cold phosphate buffer saline (PBS), mixed with sodium dodecyl sulfate sample (SDS) buffer and boiled for 5 minutes. Then, the proteins were separated by electrophoresis with SDS-Page on 10% polyacrylamide gels (Bio-Rad Laboratories) and transferred to nitrocellulose membranes. After blocking with 5% non-fat milk/tris-buffer saline with tween-20 (TBST), immunoblotting was conducted with anti IRS-1 rabbit and phosphotyrosine antibody (rabbit polyclonal, 1:600 in TBST, Santa Cruz Biotechnology) to determine the changes in IRS-1 level and insulin-mediated tyrosine phosphorylation of the IR and IRS-1, respectively.

AKt/PKB in both serine 473 and threonine 308 kinases phosphorylation were electrophoresed and immunoblotted with rabbit antibodies Akt/PKB both serine 473 and threonine 308. Examination of the levels of IR and Akt/PKB protein was conducted with homogenates prepared from another set of three brain slices. Both proteins were resolved by the immunoprecipitation and immunoblot assay conducted with rabbit anti-IR and rabbit anti-Akt/PKB (1:1,000 in TBST, Santa Cruz Biotechnology). For loading control, immunoblotting for each membrane was completed incubation with anti- β-actin (1:400; rabbit polyclonal; Sigma, Missouri, USA).

All membranes for visualizing the phosphorylation and the protein levels of IR, IRS-1 and Akt/PKB were incubated with secondary goat anti-rabbit antibody conjugated with horseradish peroxidase (1:8,000 in TBST, Bio-Rad Laboratories). The protein bands were visualized on Amersham hyperfilm ECL (GE Healthcare, Buckinghamshire, UK) using Amersham ECL western blotting detection reagents (GE Healthcare). Band intensities were quantified by Scion Image and the results were shown in average signal intensity (arbitrary) units.

nNOS Immunohistochemistry

After the end of 4-, 8- and 12-week of both dietary treatment (n = 6/ group), all animals were deeply anesthetized with an intraperitoneal injection of pentobarbital (80mg/kg body weight) and perfused through the heart with cold normal saline solution (0.9%) and then the 400 ml of 4% paraformaldehyde in 0.1 M phosphate buffer (PBS), pH 7.4. Brains were dissected out of the skull, postfixed for 2 hours in the same fixative and then placed overnight in 30% sucrose solution in PBS until sectioning. The hippocampal areas of the brain were cut in the coronal plane at 40 mm thickness with the freezing microtome for nNOS immunohistochemistry. The free-floating sections were washed in PBS for 30 minutes and then were incubated for 60 minutes with 10% normal goat serum (Vector Laboratories, Burlingame, CA, USA) and incubated overnight at 4oC with nNOS rabbit antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) diluted 1: 400 in PBS. A biotinylated goat anti-rabbit secondary antibody (Vector Laboratories) was then used at a dilution of 1:200 for 60 minutes at room temperature. The antigen-antibody reaction was revealed with avidin-peroxidase complex (Vectastain ABC kit Elite, Vector Laboratories) for 60 minutes. The peroxidase activity was visualized with 3,3-diamino-benzidine (Sigma, St. Louis, MO, USA). All sections were mounted on gelatin-coated slides, air dried, cleared in xylene and cover slipped with Permount. We have also performed the controls in our materials, in which the primary antibody was omitted and replaced with an equivalent concentration of normal goat serum. We selected at least three slides for each CA1 hippocampal area of all animals and counted all the nNOS-immunoractive neurons per 100 mm2 in each corresponding area of CA1 hippocampus.

Statistical analysis

Data were presented as means ± SE. All statistical analyses were performed using the statistical program SPSS (version 16; SPSS, Chicago, Ill., USA). The significance of the difference between the means was calculated by Student's t-test with p<0.05. Simple correlation analysis was used to determine the relationship between the plasma parameters, liver triglyceride content, visceral fat, body weight and the percentages of insulin-mediated LTD.


Peripheral insulin resistance found in 12-week high-fat-fed rats

Initial animal body weight was not different among experimental groups. Animals fed with high-fat diet (HF) for four weeks had an increased in body weight compared to rats fed with normal-diet (ND) (Table 1). However, the level of plasma glucose, plasma triglyceride, plasma insulin, liver triglyceride content and HOMA index did not differ between both dietary groups.

Animals on the 8-week HF diet also had a significant increase in body weight compared to those with ND feeding (Table 1). Fasting plasma glucose levels in the 8-week HF-fed animals were significantly higher than those in the 8-week ND-fed group. In contrast to plasma glucose level, plasma insulin level, plasma triglyceride level, liver triglyceride content and HOMA index were not significantly different between both dietary groups (Table 1).

In the rats on the 12-week HF diet, body weight, visceral fat, plasma insulin level, liver triglyceride content and HOMA index were significantly increased compared to those on the normal diet (p<0.05; Table 1). However, there was no significant difference in fasting plasma glucose and triglyceride levels between both groups. The glucose responses during the oral glucose tolerance test (OGTT) of the 12-week animals of both dietary groups were used to confirm the peripheral insulin resistance following 12 weeks of HF diet feeding. The glucose response to the oral glucose load was markedly increased in 12-week HF diet group at 15-, 30- and 60 minutes time points compared to the ND group (p<0.05, Fig. 1A). The total area under the glucose curve (AUCg) was also significantly increased in the 12-week HF diet group (p<0.05, Fig. 1B). These results indicate the occurrence of peripheral insulin resistance in the rats on the 12-week HF diet. Rats fed with the HF diet consumed less food per day than those on the normal diet (18.97 + 0.45 and 20.51 + 0.45 g/day in HF and ND groups, respectively, p<0.05). However, when the consumed food weights were converted to caloric intake, rats fed with HF diet consumed more calories per day than rats fed with ND (102.05 + 2.41 and 84.10 + 1.85 g for HF and ND groups, respectively, p<0.01).

High-fat feeding for 12 weeks significantly reduced the ability of insulin to induce long term depression (LTD) of field excitatory postsynaptic potentials (fEPSPs) in hippocampal CA1 circuits.

In all time courses (4-, 8- and 12-week) of ND-fed rats, we found that insulin application to hippocampal slices reduced the size of the fEPSP responses at 2-3 minutes after the start of insulin infusion with maximum effects appearing over the following 10-15 minutes, and the depression of fEPSPs was prominent and long lasted for 30-40 minutes (Fig. 2 and 3). In our set-up, 500 nM insulin added to the infusion line required approximately 1 minute to reach the slices.

In HF animals, the degree of insulin-mediated LTD observed from slices of 4-week and 8-week HF animals was not significantly reduced compared to the 4-week and 8-week ND animals (Fig. 2). At 30-minute post-insulin stimulation, the percentage reduction of the normalized fEPSP slope from 4-week ND and 8-week ND groups were 78.57 + 6.92 % and 72.19 + 4.74 % of the average slope recording during the baseline level, respectively (n=7-8 independent slices per group, n=6 animals/group). The percentage reduction of fEPSPs of 4-week and 8-week HF slices were 74.86 + 5.98% and 76.88 + 3.84 % of the values recorded before insulin application, respectively (n=7-8 independent slices per group, n=6 animals/group) (Fig. 2).

In 12-week HF-fed group, the amount of insulin-mediated LTD was significantly diminished (p<0.05 vs. ND group, Fig. 3). At 30-minute post-insulin stimulation, the percentage reduction of the normalized fEPSP slope from 12-week ND was 73.60 + 4.18 % of the average slope recording during the baseline level (n=16 independent slices, n=14 animals/group), while the percentage reduction of fEPSPs of 12-week HF slices was 9.34 + 3.09 % of the values recorded before insulin application (n=17 independent slices, n=13 animals/group) (Fig. 3). These results indicated that 12-week HF diet feeding caused the impairment of neuronal insulin receptor responses by suppressing the effect of insulin-mediated LTD. Furthermore, we found that the average degree of insulin-mediated LTD at all time courses of both dietary groups was significantly correlated with several peripheral insulin resistance parameters, such as body weight (r = -0.717), visceral fat (r = -0.883), plasma insulin level (r = -0.440) and liver triglyceride content (r = -0.6.15) (p<0.01).

We found no significant difference in the pair-pulsed facilitation (PPF), which indicates the alteration of presynaptic transmission, of slices recorded at the baseline between both dietary groups at all time courses (Fig. 4A, p > 0.05). These findings indicate that the release of neurotransmitter was unchanged following neuronal insulin resistance, and that the ability of CA1 hippocampal neurons to express short-term changes was not altered in 12-week HF-fed rats. To confirm whether the impairment of the insulin-mediated LTD in 12-week high-fat diet group could occur without the interruption of the other form of synaptic plasticity, the experiment of carbachol-induced LTD was determined. We found that the degree of carbachol-induced LTD in CA1 hippocampus of both 12-week high-fat diet and normal diet groups were similar (Fig. 4B). These findings suggested that the reduction of insulin-mediated LTD in 12-week HF hippocampal slices was mainly due to neuronal insulin resistance occurring at the post-synaptic sites but not due to a non-specific alteration of pre-synaptic transmission, and that the impairment of neuronal insulin receptors did not disrupt other form of synaptic plasticity.

Phosphorylation of insulin receptor (IR), insulin receptor substrate- 1 (IRS-1) and Akt/PKB levels in brain slices was depressed in 12-week high-fat-fed rats.

In order to compare neuronal insulin receptor signaling within 4-, 8- and 12-week time courses between HF and ND rats, we, first, investigated whether the protein levels of IR, IRS-1 and Akt/PKB were down-regulated in relation to the reduction of the ability of insulin-mediated LTD in CA1 hippocampus. The amount of IR, IRS-1 and Akt/PKB protein was illustrated via immunoprecipitation and immunoblotting assays. We found that the levels of the IR, IRS-1 and Akt/PKB from 4-, 8- and 12-week HF brain slices were not significantly different from 4-, 8- and 12-week ND brains (Fig. 5, n=6-8 per group of rats). The densitometric quantification of blots illustrated that the IR/β-actin protein level in the 4-week HF group was 1.16 ± 0.22 and in the 4-week ND group was 1.26 ± 0.19 arbitrary scanning units (p=0.74, n=4/group, Fig. 5A). The IR/β-actin protein level in the 8-week HF group was1.22 ± 0.07 and in the 8-week ND group was 1.23 ± 0.08 arbitrary scanning units (p=0.9, n=6/group, Fig. 5A). The IR/β-actin protein level in the 12-week HF group was 1.32 ± 0.09 and in the 12-week ND group was 1.38 ± 0.06 arbitrary scanning units (p=0.6, n=8/group, Fig. 5A).

The IRS1 protein level in the 4-week HF group was 134 ± 8 and in the 4-week ND group was 125 ± 14 arbitrary scanning units (p=0.6, n=5/group, Fig. 5B). The IRS1 protein level in the 8-week HF group was 142 ± 5 and in the 8-week ND group was 141 ±10 arbitrary scanning units (p=0.9, n=6/group, Fig. 5B). Although the IRS-1 protein levels in the 12-week HF group revealed a 20 % reduction from those slices observed in the 12-week ND group, it was not significantly different (91 ± 12 vs. 114 ± 12 arbitrary scanning units for the 12-week HF and 12-week ND groups, respectively, p=0.1, n=8/group, Fig. 5B).

The Akt/PKB/ β-actin protein level in the 4-week HF group was 1.04 + 0.05 and in the 4-week ND group was 1.07 + 0.08 (p >0.05, n=4/group, Fig. 5C). The Akt/PKB/β-actin protein levels in the 8-week HF group was1.00 + 0.02 and in the 8-week ND group was 1.03 + 0.03 (p >0.05, n=4/group). The Akt/PKB/β-actin protein levels in the 12-week HF group was 0.99 + 0.02 and in the 12-week ND group was 1.00 + 0.02 (p >0.05, n=8/group).

The phosphorylation status of IR, IRS-1 and Akt/PKB in the acutely prepared brain slices under basal condition following the insulin stimulation is illustrated in Fig. 6. The basal (non-insulin stimulated) phosphorylation levels of IR and IRS-1 from the 4-, 8- and 12-week ND and from the 4-, 8- and 12-week HF groups were weakly detected by immunoprecipitation and phosphotyrosine immunoblotting. However, insulin stimulation resulted in the strong observable phosphorylation of both IR, IRS-1and Akt/PKB levels in 4-, 8- and 12-week ND and 12-week HF groups (Figs. 6A-D). In 4-, 8- and 12-week ND-fed rats, the exposure to insulin stimulation resulted in an increase in IR, IRS-1 and Akt/PKB phosphorylation, compared to their own basal condition. However, only in the 12-week HF group that the insulin stimulation did not alter IR, IRS-1 nor Akt/PKB (serine 473 and threonine 308) phosphorylation, compared to their own basal condition. Interestingly, at 12-week feeding, the insulin-stimulated IR phosphorylation was significantly lower in the HF group (0.96±0.03) than in the ND group (1.28±0.06), accounting for ~ 25% reduction (p=0.00, n=11/group, Fig. 6A). Similarly, the insulin-stimulated IRS-1 phosphorylation was significantly lower in the 12-week HF group (1.02±0.04) than in the 12-week ND group (1.20±0.03), accounting for ~ 15% reduction (p=0.00, n=10/group, Fig. 6B). Furthermore, since the serine/threonine kinase Akt/PKB is one of main insulin signaling pathways and plays role in the mediating of hormone's metabolic effect (43), and that the phosphorylation levels of the two residues is necessary for Akt/PKB activity, we determined the levels of serine 473 and threonine 308 in the present study. From the brain slices of the 12-week HF group, insulin significantly impaired the ability of insulin to cause Akt/PKB phosphorylation of serine 473 and threonine 308 (Fig. 6C and 6D). These findings indicate that the integrity of neuronal insulin receptor signaling, the phosphorylation of IR, IRS-1 and Akt/PKB in the 12-week HF-fed rats, and the ability of insulin-mediated LTD in CA1 hippocampus were significantly reduced.

The number of nNOS-positive neurons was significantly reduced in 12-week high-fat diet group.

Insulin signaling pathways: IR, IRS-1 and Akt/PKB, have been shown to play important roles in neuronal survival (44). In addition, previous studies demonstrated that insulin resistance can accelerate ageing syndrome (45), and that the number of nNOS immunoreactive neurons in cerebral cortex and hippocampus was significantly decreased in the brain of aged rats (46). In the present study, we found that 12-week HF consumption significantly reduced neuronal insulin signaling and particularly decreased insulin-mediated Akt/PKB phosphorylation, indicating neuronal insulin resistance. Thus, we further hypothesized that rats fed with 12-week HF diet have increased neuronal ageing, compared to 12-week ND group. To test this hypothesis, we used nNOS-immunoreactivity as a neuronal ageing marker to investigate whether rats with neuronal insulin resistance caused by HF diet feeding for 12 weeks would develop neuronal ageing in CA1 hippocampus faster than rats fed with ND diet.

In the present study, we found that nNOS-immunoreactive neurons in all animals were observed in the interneurons of CA1 hippocampus (Fig. 7A). There was no significant difference in the number of nNOS-immunoreactive neurons between HF diet and ND groups following 4- and 8- week of feeding (Fig. 7B). However, the number of nNOS-immunoreactive neurons in CA1 hippocampus of 12-week HF group (19.9 + 0.8 neurons/ 100 mm2, n=22 brain slices/6 animals) were significantly decreased compared to 12-week ND group (23.1+ 0.86 neurons/ 100 mm2, n=29 brain slices/6 animals; p<0.05, Fig. 7B). These data suggest that neuronal insulin resistance could accelerate neuronal ageing in CA1 hippocampus.


Insulin resistance is quickly becoming one of the world's most prevalent metabolic disorders (47-49). Earlier works found that HF diet causes insulin resistance characterized by hyperinsulinemia, hyperlipidemia and decreased insulin sensitivity (1, 50). While considerable research has examined both the consequences and mechanisms of a diminished insulin response in various peripheral tissues, only a few studies have investigated the effects of this metabolic disruption within the CNS, particularly the metabolic disturbance following the consumption of HF diets. In humans and rodents, it has been shown that the development of insulin resistance is associated with HF consumption and is linked to cognitive deficits (1, 6, 7). Although much evidence suggests that neuronal insulin signaling might play a role in neuronal plasticity (4, 51, 52), the number of available reports in this area is still limited.

Growing evidence demonstrates the influence of time-course effects on the learning and memory processes caused by HF diet. Previous studies demonstrated that rats fed with HF diet for 3 months had cognitive impairment (53), while rats fed with HF diet for 8 months exhibited impaired learning ability and reduced hippocampal synaptic plasticity (34). Despite these reports, the neurofunctional insulin resistance in hippocampus in different time-course of HF consumption has not been investigated. To determine the existence of neurofunctional insulin resistance in the brain, we determined the efficacy of the neurofunctional insulin receptor characterized as insulin-mediated LTD in CA1 hippocampus and the stimulated phosphorylation status of IR , IRS-1 and Akt/PKB in brain slices harvested from control and HF fed rats. We demonstrated a significant reduction of the insulin-mediated LTD in CA1 hippocampal slices and the insulin-mediated phosphorylation of IR, IRS-1 and Akt/PKB in 12-week HF-fed rats. Our findings suggest that the consumption of HF diet for only 12 weeks can down-regulate the neuronal insulin receptor sensitivity as well as the peripheral insulin sensitivity. A previous study on hamsters with peripheral insulin resistance resulting from a 6-week high-fructose diet has shown reduction of IR, IRS-1 and Akt/PKB phosphorylation in the brain and insulin-mediated LTD in hippocampus (32). Our study demonstrates that the HF diet model required a longer time than did fructose diet model to observe the neuronal insulin resistance.

In the present study, insulin resistance, characterized by weight gain, increased visceral fat, hyperinsulinemia, increased HOMA index and increased OGTT were observed following the 12-week HF diet. However, we found that the fasting plasma triglyceride level following consumption of the 12-week HF diet was not changed, whereas the level of liver triglyceride in the 4-, 8- and 12-week HF-fed rats significantly increased, compared to the 4-, 8- and 12 week normal diet. This finding is similar to that reported previously, in which HF diet-induced obesity in rats increased liver triglyceride content without any change in plasma triglyceride level (54). A possible explanation could be that increased ingestion of fats preferentially contributed towards the cytosolic pool by increasing liver triglyceride concentration and longer time is required for the up-regulation and secretion of plasma triglyceride. Furthermore, excessive intake of fat leads to an accumulation of triglyceride in many tissues, particularly in the adipose tissue (50). Supporting this is the finding that rats fed with 4-, 8- and 12-week HF diet have dramatically increased visceral fat. In addition, our data demonstrated that the plasma insulin level in 12-week HF group significantly increased without any changes in plasma glucose level, suggesting that rats with 12-week HF diet feeding were in the peripheral insulin resistance status and the unchanged plasma glucose level was the result of the compensating mechanism of hyperinsulinemia.

Long-term depression (LTD) is a phenomenon that reflects neuronal adaptation and is thought to provide a functional measurement of synaptic plasticity (55). Insulin receptors are dispersed throughout the brain with the highest density located in the hippocampus, where these receptors may regulate glucose homeostasis and brain function such as learning and memory (52, 56, 57). Insulin has been shown to regulate the endocytosis of AMPA receptors, which causes the depression of excitatory synaptic transmission (26-29, 58).

In the present study, a significant reduction of insulin-mediated LTD in hippocampal slices from the 12-week HF group was observed. The reduction of insulin-mediated LTD was well correlated with other peripheral insulin resistance such as visceral fat, weight, plasma insulin level and liver triglyceride content. These results suggest that changes in metabolic system are linked to the neuronal function. The weakening of insulin-mediated LTD indicated one of the functional consequences of impaired neuronal insulin signaling. The reduced LTD induction could be due to the significant decrease in insulin-induced tyrosine phosphorylation of the insulin receptor signaling: IR, IRS-1 and Akt/PKB in brain slices of the 12-week HF group (Fig. 6). Although it is possible that the diminished insulin-mediated LTD in the HF group could be due to changes in synaptic transmission in CA1 hippocampus caused by HF, the unaltered PPF hippocampal CA1 regions in both dietary treatments suggested that HF feeding did not affect presynaptic responses to electrical stimulation. Therefore, it may be concluded that 12 weeks of HF feeding, which induced peripheral insulin resistance, only significantly affects the neuronal insulin receptor signaling function at the post-synaptic sites, but not at the pre-synaptic release in CA1 hippocampus. Carbachol-induced LTD or muscarinic LTD (mLTD) in CA1 hippocampus, characterized as the other form of synaptic plasticity (59), is believed to be substrates of learning and memory at the molecular level (59). We demonstrated in this study that the reduction of neuronal insulin response in the 12-week HF group was unaffected in mLTD, suggesting that this form of synaptic plasticity may not require neuronal insulin signaling. Our finding was similar with a previous study, showing that the weakening of insulin-mediated LTD has no effect upon the induction and maintenance LTP via high-frequency stimulation (32).

In the present study, the phosphorylation of IR, IRS-1 and Akt/PKB in brain slices was diminished, whereas the levels of IR, IRS-1 and Akt/PKB in brain slices were not altered by 4-,8- and 12-week HF diet feeding (see Fig. 5). These changes in phosphorylation of neuronal insulin signaling confirm the impairment of insulin receptor function in the brain following 12-week HF feeding. The unchanged levels of IR, IRS-1 and Akt/PKB proteins following HF diet consumption in the present study are consistent with those in previous reports, in liver (60), skeletal muscle (61) and hippocampus (62). Furthermore, the impairment of IR, IRS-1 and Akt/PKB tyrosine kinase activity has been demonstrated in skeletal muscle (63), fat (64) and liver tissues (65) in 10-12 week consumption of HF diets. All of these findings suggest that HF diet could cause defective neuronal insulin receptor function, but not at the level of protein expression.

In addition to the weakening of neurofunctional insulin receptors and neuronal insulin signaling, we also found that 12-wk HF-fed rats had the reduction of nNOS expression in hippocampus. Neuronal NOS expression in cerebral cortex and hippocampus has been shown to correlate with ageing in animals (66). Therefore, our findings suggest that neuronal insulin resistance following HF diet consumption may lead to neuronal ageing similar to ageing syndrome occurring following peripheral insulin resistance (67).

In summary, the present study demonstrates that a rapid significant modification of important neuronal insulin receptor signaling can be induced by a fat-enriched diet. Fed for 12 weeks, the HF diet clearly induces neuronal insulin resistance, which is identified as a significant reduction in the ability of insulin to induce LTD, and a reduction in the stimulated phosphotyrosine activity of IR, IRS-1 and AKt/PKB in brain slices. Twelve-week HF feeding not only causes neuronal insulin resistance, but also leads to neuronal ageing. Since the defective insulin receptor signaling has been shown to associate with the pathogenesis of Alzheimer's disease, (18), cognitive impairment (6, 7, 53, 68) and the presence of cognitive impairment in patients with type II diabetes (69), the neuronal insulin resistance developing after 12-week HF consumption could be responsible for to the impairment of cognition in this animal model.


The authors wish to thank Prof. M. Kevin O' Carroll, Professor Emeritus, University of Mississippi School of Dentistry, USA, and Faculty Consultant, Faculty of Dentistry, Chiang Mai University, Thailand, for his editorial assistance. This work is supported by the Thailand Research Fund grants: TRF-RMU5180007 (SC), TRF-RTA5280006 (NC) and the Faculty of Medicine Endowment Fund, Chiang Mai University (WP, AP, NC and SC).