Insulin effects are predominant in skeletal muscles, liver, kidney, fat and brain causing increased renal sodium retention, modulation of transmembrane cation transport, induction of growth promoting effects of vascular smooth muscle cells and vascular hyperreactivity 21. Insulin actions are initiated when insulin binds to a high-affinity heterotetrameric transmembrane protein receptor that is present in all mammalian cells 22. This is followed by activation of second messengers, also called docking proteins, such as the insulin receptor substrates-1, -2, -3 and -4, (IRS-1, -2, -3 and -4) 23242526 via a series of phosphorylation-dephosphorylation reactions that result in stimulation of glucose metabolism 27 (Figure 2). While IRS-3 and -4 play a role in cell growth and differentiation, IRS-1 and -2 play an important role in glucose metabolism and represent attractive candidate genes to study in insulin resistance (Tables 1 and 2).

In humans, associations between insulin resistance and common variants in IRS-1 and -2 have been reported in several populations 25282930313233343536, including obese Caucasian children, adults, Asian Indians, Mexicans and Europeans. Mechanisms underlying the contribution of IRS-1 and/or -2 variants to insulin resistance include 37 (Figure 3): i) altering IRS-1 and/or-2 expression and function, ii) reduced IRS-1 and/or -2 binding to the insulin receptor, iii) a defect in binding of IRS-1 and/or -2 variant (s) to the p85 regulatory subunit of the PI3-kinase and a decrease in PI3-kinase activity. This in turn leads to either a decreased GLUT4 translocation to the plasma membrane, further reducing glucose transport and glycogen synthesis, or a significant IRS-1 and/or -2 -induced decrease in phosphorylation of glycogen synthase kinase-3 (GSK-3), an enzyme that is important in glycogen synthesis, thus causing reduced glycogen synthesis, iv) reduced IRS-1 content that is not compensated by a constitutive increase in the IRS-2 protein content. This result in a reduced insulin-stimulated PI3-kinase activity and a significant decrease in Akt phosphorylation and activity. Overall, IRS-1 and/or 2 variants seem to impair the ability of insulin to activate the IRS/PI3-kinase/Akt/GSK-3 signaling pathway leading to defects in glucose transport, glucose transporters translocation and glycogen synthesis.

IRS-1 and -2 proteins are subjected to phosphorylation on either tyrosine or serine residues (Figure 4 for structural organization). IRS-1 and -2 phosphorylation on tyrosine residues activates IRS proteins to bind to signaling molecules containing SH2 domains, including PI3K 2438. Unlike tyrosyl phosphorylation, serine phosphorylation of IRS proteins attenuated insulin signaling and might potentially explain an additional mechanism of insulin resistance in rodents 39 (Figure 5). Owing to the fact that IRS proteins have three times the number of serine residues compared to the number of tyrosine residues, the significance of serine phosphorylation has been significantly highlighted. Serine phosphorylation of IRS-1 and IRS-2 is the main mechanism of suppressing insulin signaling 4041 in the following ways 404142: 1) They can induce dissociation of IRS proteins from the insulin receptor, 2) hinder tyrosine phosphorylation sites and release IRS proteins from intracellular complexes that maintain them in close proximity to the receptor, 3) induce IRS protein degradation, or 4) turn IRS proteins into inhibitors of the insulin receptor kinase. Differential serine phosphorylation on IRS-1 and/or-2 has not been studied in Dahl S versus R rats. Additionally, serine phosphorylation of the insulin receptor and IRS-1 and -2 43 is enhanced by chronic hyperinsulinemia that negatively modulate insulin signaling via activation of TNF-α 43. Therefore, chronic hyperinsulinemia in high-salt-fed-Dahl S rats 16 might as well enhance serine phosphorylation on IRS-1 and/or IRS-2 and explain insulin resistance in Dahl S rats.

The ability to transport glucose across plasma membranes is mediated by members of the glucose transporter proteins (Table 3). Two families of glucose transporter have been identified: i) GLUTs: facilitated-diffusion glucose transporter family also known as 'uniporters,' (GLUT1 through GLUT4), ii) SGLTs: sodium-dependent glucose transporter family, also known as 'cotransporters' or 'symporters', transport glucose against its concentration gradient.

In humans, 23, 38, 2, 1 and 41 mutations have been reported in GLUT 1, GLUT 2, GLUT 4, GLUT 10, and the SGLT 1, respectively. These mutations have been associated with glucose transporters deficiency, lower serum insulin levels, glucose malabsorption and NIDDM making them attractive molecular targets for insulin resistance and glucose abnormalities 44 (For review, the Human Gene Mutation Database).

In rats, at least 7 genes are involved in glucose transport. They are denoted by Slc2a1, Slc2a2, Slc2a3, Slc2a4, Slc2a8, Slc37a4, Naglt1 (Table 3). Despite the reported variations in some of the above genes in rats (Table 4), no further functional studies have been reported so far. Using the TRANSFAC ^® computer program 45, sodium-binding sequences 1314 were identified in the promoters of Slc2a1, Slc2a2, and Slc2a4. As such, association between high salt and insulin resistance might be through the glucose transporter genes.

Inflammation causes a profound change in gene expression of a large number of inflammatory proteins such as tumor necrosis factor-alpha (TNF-α), c-Jun Terminal Kinase (JNK) and nuclear factor-kappa B, (NF-kappa B) which increases oxygen consumption and produce many reactive oxygen species (ROS). ROS might as well be triggered by high salt 46. ROS interfere with normal insulin signaling and precipitate insulin resistance via a number of mechanisms including: i) disruption of insulin-induced IRS-1 and PI3-kinase phosphorylation and cellular distribution, ii) reduction in GLUT4 expression levels, or iii) reduction in GLUT4 translocation to the plasma membrane 47484950. High salt-fed Dahl S rats had augmented NADPH oxidase activity (oxidative stress inducer) in cardiac 10, renal 91112, mesenteric microvascular 9 and brain tissues 51. Even plasma levels of hydrogen peroxide (oxidative-stress inducer) in high-salt-fed Dahl S rats were significantly elevated 9, and the renal antioxidant, superoxide dismutase levels were diminished 9. As such, high salt-induced oxidative stress might predict a powerful inflammatory response in Dahl S versus R rats because of genetic differences along the inflammatory pathway that predispose them to perturbations in the expression of inflammatory genes.

Table 5 includes a list of inflammatory genes (pro- and anti-inflammatory factors) with potential implications in insulin resistance. While there are a plethora of genes associated with inflammation and insulin resistance, I will focus on a select few TNF-α, JNK, NF-kappa B which based on the literature are plausible candidates. TNF-α and JNK are potential mediators of insulin resistance and are significantly upregulated in high-salt-fed-Dahl S rats 5253. TNF-α and JNK contribute to insulin resistance by: i) promoting serine phosphorylation of IRS-1 and -2, ii) formation of a stable complex with IRS-1 and/or IRS-2, iii) impairing the ability of IRS-1 and -2 to associate with the insulin receptor and inhibiting insulin-stimulated tyrosine phosphorylation 54555657.

On the other hand, high salt-fed Dahl S rats possess high renal nuclear factor-kappa B, NF-kappa B 58, which might predispose them to insulin resistance. This is because inhibition of the nuclear factor I-kappa B kinase B (IKBKB), which is a member of the NF-kappa B, with salicylates or through targeted gene disruption causes a dramatic improvement of insulin sensitivity in animal models of insulin resistance such as ob/ob mice and obese Zucker fatty rats 5960. It also reversed hyperglycemia, hyperinsulinemia, and dyslipidemia in obese rodents. On the other hand, activation or overexpression of the nuclear factor I-kappa B (IKBKB) attenuated insulin signaling in cultured cells 60.

As such, common variant (s) in the TNF-α, JNK, or NF-kappa B might predict a powerful inflammatory reaction in response to high salt and might possibly explain insulin resistance in Dahl S rats.

Glucose utilization requires three steps: i) delivery of glucose to the plasma membrane, which depends on the organ's blood flow, capillary recruitment and endothelial permeability to glucose 61, ii) facilitated transport across the membrane, which mainly relies on the glucose transporters numbers and intrinsic activity 61 and iii) intracellular phosphorylation to glucose-6-phosphate by a hexokinase cytosolic enzyme (HK) (Figure 6) 62.

In most cases, glucose transport is believed to be the rate limiting step. However, in conditions of hyperinsulinemia like that seen in high-salt-fed Dahl S rats, the rate limiting step may switch to phosphorylation 6364. Perturbations in any of these steps might precipitate the high salt-induced insulin resistance in Dahl S rats. Variations in the genes proposed in the current study might affect any of the three steps involved in glucose utilization. As for the hexokinase enzyme, increased adrenal content of norepinephrine and epinephrine in high-salt-fed Dahl S versus R rats might contribute to a defective hexokinase enzymatic activity in Dahl S rats 65. Epinephrine decreases glucose clearance in vivo upon insulin stimulation 6667 and suppresses insulin-stimulated glucose uptake in skeletal muscles 686970. This is possibly by: i) stimulating glycogenolysis and accumulating glucose-6-phosphate 687172 (a strong inhibitor of the hexokinase enzyme) 7374; or ii) by reducing glucose transport across the membrane 757677.

Despite comparable muscle and adipose tissue GLUT4 protein levels in Dahl S versus R rats on high and normal salt diet, Dahl S rats showed reduced glucose uptake by the muscle 19. Reduced glucose uptake in high salt-fed-Dahl S rats might be explained by the high salt-induced release of epinephrine and norepinephrine that directly and indirectly inhibit glucose phosphorylation and utilization. Moreover, high salt causes solubilization of the hexokinase and subsequent inhibition 78, which might be an additive mechanism in reducing glucose uptake by peripheral tissues in Dahl S rats. As such, high salt combined with genetic differences in Dahl S versus R rats might predispose them to insulin resistance.

High salt diet combined with genetic differences along the insulin signaling and inflammatory pathways predict a powerful inflammatory response and susceptibility to high salt-induced insulin resistance in Dahl S rats.

○ Dahl S rats harbour common variant(s) in the genes encoding the insulin receptor substrates-1 and/or-2 and/or the glucose transporter proteins.

○ Dahl S rats will display a more powerful inflammatory response to high salt-induced oxidative stress because of common variant(s) in the genes encoding the tumor necrosis factor-alpha, c-jun terminal kinase and the nuclear factor-kappa B.

○ Dahl rats will show reduced glucose delivery, transport and phosphorylation in liver and kidney tissues compared to Dahl R rats. These differences will be related to the combined effect of high salt diet and genetic differences along the insulin signaling and inflammatory pathways.

To identify putative gender differences to susceptibility to high salt-induced insulin resistance, four-week-old male and female Dahl salt-sensitive (Dahl S) (n = 200) and salt-resistant (Dahl R) (n = 200) rats will be purchased from Harlan Sprague Dawley (Indianapolis, IN) and will be fed a standard rodent diet containing 0.3% NaCl (normal diet group) or 8% NaCl (high-salt group) or 0.03% NaCl (low-salt group) during a 4-week experimental period. The rats will be housed in a room maintained at constant humidity (60 ± 5%), temperature (23 ± 1°C), and light cycle (12 hours: 0700 to 1900). Food and tap water will be available ad libitum throughout the study. All experimental procedures will be carried out in accordance with the guidelines of the University of Ottawa Animal Care Committee for the care and use of laboratory animals. The rats will be characterized before and after normal, high and low salt diet in terms of: body weights, food intakes, blood pressures and plasma parameters (Fasting blood glucose, fasting plasma insulin). The blood pressure of the rats will be measured every other day by the tail-cuff method using a Softron BP-98A automatic sphygmomanometer 19. Rats will be acclimated to the procedure daily during the week leading up to study. Blood glucose and plasma will be assayed using the glucose oxidase and radioimmunoassay methods, respectively. This will be performed every week using blood samples obtained from a tail vein (1 ml) 19. Measurements of blood pressure, blood glucose and plasma insulin will continue till the fourth week post treatment. Prior to sacrificing the rats, Dahl S versus R rats will be confirmed as insulin resistant using the hyperinsulinemic-euglycemic clamp procedure 19.

Rats will be euthanized 4 weeks post treatment and genomic DNA will be isolated from white blood cells using Qiagen Human FlexiGene DNA kit according to the instructions provided by the manufacturer. 5 mls of blood/rat are sufficient to produce ample DNA for our study. Kidney and liver tissues will be collected and immediately flash frozen in liquid nitrogen for mRNA and protein extraction. Kidney and liver tissues are good candidate tissues owing to their importance in the pathogenesis of insulin resistance. Abnormal responses of insulin receptor mRNA levels to high salt intake were reported in kidneys and livers of animal models of insulin resistance such as the spontaneously hypertensive rats (SHR) and the fructose-fed rats 7980.

Variants currently identified in genes encoding the IRS-1, IRS-2 and GLUTs proteins correlate with insulin resistance. In rats (Sprague Dawley), only one non synonymous polymorphism has been reported in the IRS-1 gene so far. However, functional studies are yet to be determined (Tables 1 and 2). Two independent methods of genotyping namely DHPLC and sequencing will be combined as was done previously 81828384. DHPLC offers sensitivity of 95%–100% and specificity of 100% 85, while sequencing offers sensitivity of 99.7%–100% and specificity of 100% respectively 86. Combining these highly sensitive and specific methods of genetic screening will help eliminate any false positive or false negative results that might occur with one method of genetic screening. Sequences will be retrieved from the GenBank [GenBank: IRS-1 NM_012969;IRS-2: XM_001076309]. Dahl S and R rats' sequences will be compared to each other and to the Rattus Norvegicus (Brown Norway rat) sequence readily available in the Rat Genome Database. Brown Norway rat sequence will serve as an additional control.

Aim 2: Determination of mRNA and protein expression levels for genes encoding IRS-1, -2 and glucose transporter proteinsm

Kidney, liver and brain tissues will be utilized for total RNA extraction using the Qiagen RNeasy Mini Kit 878889. Fold differences in mRNA levels for genes encoding IRS-1, -2 and the GLUTs glucose transporter proteins will be obtained using quantitative RT-PCR 90.

ReadyPrep™ Protein Extraction Kit (Signal), from Bio-Rad, will be utilized to extract total protein from the kidney, liver and brain tissues of Dahl rats after 4 weeks. The availability of an antiserum specific for the IRS-1 and -2 and GLUT proteins (Upstate Biotechnology Inc. Lake Placid, NY) will allow us to probe for these proteins in kidney, liver and brain cell homogenates.

Aim 3: Are IRS-1 and -2 differentially phosphorylated on serine residues

Variations in genes encoding IRS-1 or -2 might be at the level of DNA (genetic variants) or at the level of serine phosphorylation. In order to examine differential serine phosphorylation of IRS-1 and -2 in Dahl S, versus R rats, anti-IRS-1 phosphoserine antibody will be utilized to probe kidney, liver and brain samples from low, regular and high salt diet. Fold increase or decrease in IRS-1 and -2 serine phosphorylation in Dahl S rats post dietary treatment will be compared with control Dahl R rats.

Ogihara et al., in 2002 reported an enhanced tyrosine phosphorylation of the insulin receptor, insulin receptor substrates, PI3K activation and serine phosphorylation of Akt in Dahl S versus R rats 19. However, there were some limitations to the procedures they used. First, Dahl rats were fasted for 12 hours prior to the experiment assessing IRS-1 and -2 phosphorylation. Fasting is reported to enhance insulin-induced IRS-1 and IRS-2 tyrosine phosphorylation and association with PI3-kinase in liver and muscle of animal models of insulin resistance (streptozotocin treated rats) 91. Secondly, the authors injected insulin in the portal vein of rats once after which immediately they harvested livers and hindlimb muscles for measuring the IRS-1 and -2 levels and tyrosyl phosphorylation. Co-immunoprecipitation experiments of IRS-1 and -2 were done in triplicates. Regardless of the fact that the authors has done the insulin injection once and then repeated the immunoblotting experiment three times on the harvested liver and hindlimbs of Dahl rats, it is documented that within 1 minute of acute insulin stimulation, enhancement of the insulin signaling pathway and augmentation of IRS content and tyrosyl phosphorylation together with PI3K activation occur 9293. On the contrary, chronic effects of insulin have been shown to enhance IRS-1 serine phosphorylation preventing its subsequent tyrosine phosphorylation 94. Therefore fasting and acute insulin injection in Dahl S rats (a model of insulin resistance) were additive in enhancing the insulin signaling pathway. It would have been more proper to show the impact of chronic insulin stimulation in high-salt-fed Dahl S rats without fasting or acute insulin stimulation, both of which might have misrepresented the actual defects in insulin signaling in Dahl S rats. Therefore, I propose to use the same procedure adopted by Sechi et .al., 1997, where the animals were neither fasted nor injected insulin prior to the experiment 95.

It is also possible that even in the presence of enhanced IRS-1 and/or -2 tyrosyl phosphorylation in Dahl S rats that enhanced serine phosphorylation in Dahl S rats would be the mechanism behind high salt-induced insulin resistance in this model. Serine phosphorylation might induce some conformational change that does not affect tyrosine phosphorylation but inhibits the insulin signaling pathway, via rapid tyrosyl dephosphorylation and deactivation or by acting as inhibitors to the insulin receptor kinase 404142.

High salt might as well exert an additional inhibitory effect on IRS-1 and -2, possibly by interacting with a sodium-binding sequence 1314 on the IRS-1 and/or -2 genes to suppress their mRNA and/or protein levels in Dahl S versus R rats. I was able to locate a sodium response element in the proximal promoters of IRS-1 and IRS-2 genes. As such, associations between high salt intake and insulin resistance might exist through the IRS-1 and/or -2 genes.

Aim 4: Determination of the impact of any variant/variants on the activity of IRS-1 and IRS-2, GLUTs

To test the impact of identified 5' flanking region variant(s) on the activity of IRS-1, IRS-2 and GLUTs, reporter genes can be used to assay for the activity of the promotor in a cell. The reporter gene (e.g. luciferase or GFP) is placed under the control of the IRS-1 or IRS-2 or GLUT promotor and the reporter gene product's activity is quantitatively measured. The results are normally reported relative to the activity under a "consensus" promoter known to induce strong gene expression. For variants identified in the coding region, we will assess the impact of such variant in altering the resultant amino acid sequence. The IRS-1 and IRS-2 constructs will be prepared using the same techniques we previously optimized 878890. Briefly, RT-PCR will be employed to amplify full length genes with the variant (s) identified from Dahl S rats' cDNA. Full length constructs from Dahl R rats' cDNA will be used as a control. Representative kidney and liver cell lines such as COS-7 cells and HepG2 cells respectively will be transfected with either the mutant or the corresponding wildtype as control. mRNA and protein levels will be further assessed by QRT-PCR and western blotting respectively to test the impact of such variant (s) in altering the mRNA and/or protein levels. Co-transfection studies will be performed to investigate the impact of combined variants in altering the mRNA and protein expression of IRS-1 and IRS-2, and whether a synergistic or an inhibitory response will be the outcome of such combination.

Question 2:

Aim 1: Determination of expression levels of candidate genes along the inflammatory pathway

The somewhat disappointing results of linkage studies for multi-factorial phenotypes coupled with the theory that an individual phenotype results from the cumulative effects of several contributing loci, has lead genetic investigators back to the study of candidate genes in case-control designs 96. Identification of such candidates has become increasingly rapid with modern molecular methods such as microarrays. High-throughput microarray platforms are capable of measuring the expression of thousands of genes simultaneously allowing for the identification of genes that are differentially expressed in tissues from various individuals. Despite the recognized limitations (variation in expression resulting from polymorphism, environmental variation, or RNA quality) they can potentially reveal novel molecular mechanisms contributing to phenotypes as well as uncover alternate candidate genes derived without previous bias. As such, cDNA microarrays offers:i) detection of previously unsuspected genes with a likely role in inflammation, ii) follow-up of much over and underexpresion of gene populations compared with controls, iii) identification of phenotypically-associated specific patterns of gene expression.

Before pursuing genes along the inflammatory pathway (TNF-α, JNK and NF-kappa B ) (Table 5), fold differences in mRNA levels will be obtained in microarray experiments and confirmed using quantitative RT-PCR. Hence haplotype analyses will only be performed on those genes whose mRNA expression patterns, protein levels or activity, significantly differ within or between the Dahl S and R population pre and post-high salt diet. Despite the high cost associated with the use of microarrays, they will enable us to identify potential inflammatory genes whose expression might be altered by diet in Dahl S versus R rats, and might likely influence insulin resistance in Dahl rats. Microarrays will also eliminate the most time consuming and expensive component of a future candidate gene study, and will simply direct our genetic focus to potential and differentially expressed candidate genes.

Aim 2: Searching for polymorphisms in candidate genes along the inflammatory pathway (Methodology Section-Question 1 Aim 1).

Aim 3: Determination of the impact of any variant/variants on the activity of TNF-α, JNK and NF-kappa B (Methodology Section-Question 1 Aim 4)

Aim 4: Construction of SNPs haplotypes

For haplotype construction, previously reported polymorphisms (mostly SNPs) will be identified for each candidate gene by a careful screening of available polymorphism databases. This approach is enhanced by the completion of the rat genome sequence by the Rat Genome Sequencing Project Consortium and the rapidly growing databases of rat polymorphisms, including those of The Rat SNP Consortium. Furthermore, in order to identify potential mutations and/or new polymorphisms, a systematic screening of the proximal promoters (2500 bp) will be performed. In order to exhaustively explore genetic variations, it will be essential to combine DHPLC and sequencing (ABI 310/3100 sequencers) screenings (Methodology section Question 1). DNA from rats in each population group will be used to define the haplotypes in the genes associated with differing expression post-dietary tratemnt. Haplotypes will be inferred using the PHASE software package 97. Highly efficient genotyping techniques, depending on the selected variant, will be used to genotype the populations (Denaturing High Performance Liquid Chromatography (DHPLC), Allele Specific Oligonucleoprobe (ASO), SNaPshot™ Multiplex System).

Question 3:

Aim: Determination of glucose delivery to cell membrane, glucose transport across cell membranes and glucose phosphorylation.

Measurement of interstitial glucose is relevant as it is the substrate for glucose transport. This will be performed by microdialysis as previously described 9899100. Measurement of intracellular glucose will be predictive of the efficiency of glucose transport across cell membranes. This can be done by radioactive glucose analogues that are either not transported or not phosphorylated to calculate glucose concentration localized to each side of the membrane 101102103104105. To assess perturbations in glucose phosphorylation, measuring the hexokinase enzyme activity will be performed in kidney, liver and brain tissues of Dahl rats. This will be done as previously described 106107.