Diabetes Often Shows In Families Insanity Prevails 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.

The aim of our study is to find a genetic link between schizophrenia and type-2 diabetes. The idea of this investigation comes from the fact that people who suffer from schizophrenia have more risk of developing type-2 diabetes (~9%) as compared to the general population (~ 4.9%) (Cohen et al 2006). It has also been noted that there has been high prevalence of about 30% of impaired glucose tolerance in patients having schizophrenia, subjective to age (Lindenmayer et al, 2003; Subramaniam et al, 2003). We hypothesize that the link between the two conditions is multifactorial that involves genetic and environmental, i.e., antipsychotic drug induced-obesity.

To carry the research forward we performed two genetic studies:

Association study to see if the two conditions share common genes or genetic pathway. We chose four candidate genes, AKT1, PPARG, PLA2G4A and PTGS2 in our study and the made the following 2 hypotheses:

Functional study to see the effect of clozapine (atypical antipsychotic drug) on the gene expression of various genes identified by genome wide association studies for type-2 diabetes and obesity. The hypothesis we made for this study is:

The reason of choosing PPARG gene as one of our candidate genes is because it has a role to play as a modulator of insulin sensitivity, glucose homeostasis and blood pressure. GWAS have also identified PPARG gene to be associated with type-2 diabetes. Thiazolidinediones, a group of anti-diabetic drugs that improve insulin-resistance, are identified as PPARG-ligands (Fajas et al. 1997). There is a possibility that PPARG-locus might represent a pathological link between schizophrenia and type-2 diabetes. PPARG is functionally regulated by the dehydration product of prostaglandin D2 (PGD2), 15-deoxy- prostaglandin (15-dPGJ2), which is a major endogenous ligand of PPARG.

PLA2G4A and PTGS2 genes are the other two candidate genes studied as they have been associated with schizophrenia in the past.

In this hypothesis we would test if PLA2G4A, PTGS2 and PPARG genes are associated with schizophrenia singly or in a combined manner. The results would help to determine the extent to which a genetic link explains the association between schizophrenia and type-2 diabetes.

1.1.1. Biological characters of PPARG

Peroxisome Proliferator Activated Gamma (PPAR) is a member of the PPAR subfamily of nuclear receptors. The nuclear receptor superfamily is a group of proteins that act as transcription factors to alter gene expression when bound to specific ligands. They have a unique property of forming a heterodimer complex with retinoid X receptor (RXR), another member of nuclear receptor superfamily. PPARs are involved in a wide variety of functions such as lipid metabolism, embryonic development, and virtually all endocrine pathways (Chawla A, Repa JJ, Evans RM, Mangelsdorf DJ. Nuclear Receptors and Lipid Physiology: Opening the X-Files. Science 294: 1866-1870, 2001). They also regulate a number of cellular functions from fatty acid oxidation to adipocyte differentiation. They were first identified as proteins that induce peroxisome proliferation in rodent liver cells in response to exogenous chemicals, many of which are hepatocarcinogens. There are three PPAR isotypes: , β and , which have somewhat different expression patterns and functions but have similar gene structures (Beamer, 1997; Fajas, 1997).

2.1.2. Structure of the PPARG protein

Guo & Tabrizchi (2006) proposed a structure of the PPAR protein which includes 6 structural regions (A-F) in 4 functional domains (Figure X). The details of the location and function of these domains in given below:

Ligand-independent transactivation (activation function-1) domain- AB region is transcriptionally active in the absence of ligands and can change the ligand binding affinity of the receptor when covalently modified by phosphorylation or sumoylation. It is adjacent to the NH2 terminus.

DNA binding domain - C region, is the most conserved domain in all nuclear receptors. The DNA-binding domain is formed by two zinc finger-like motifs folded in a globular structure, which can recognize a DNA target composed of 6 nucleotides (AGGTCA). This specific DNA sequence is located within the regulatory regions of the target genes and serves as the peroxisome proliferator hormone response element (PPRE), which allows binding to this domain and hence regulates transcription.

Hinge domain - D region, helps in the rotation of the DNA-binding domain and also connects to the ligand-binding domain.

Ligand-binding domain (activation function-2) - EF region, has a large binding pocket that allows the transcription receptor to bind to a variety of structure different ligands. This lies adjacent to the C terminus that is also part of the ligand-binding domain. An agonist locks ligand binding domain into an active conformation on binding whereas an antagonist stabilizes the receptor into an inactive state upon binding.

Useful links for PPARG gene:

PubMed: 7787419

OMIM ID: 601487

Entrez Gene ID: 5468

HGNC ID: 9236

1.1.3. The mode of action

PPARG becomes activated upon ligand binding and then forms a heterodimer complex with retinoid X receptor (RXR). The PPARG-RXR complex regulates the transcriptional processes by binding to a specific sequence, AGGTCA, i.e. PPREs. This sequence is separated by 1-nucleotide called DR-1 and is located on the promoter region of the target gene (IJpenberg et al. 1997). The alterations of the gene transcription depend on the nature and type of ligand binding to it. The interactions between the ligand, PPARG-RXR heterodimer, modulator proteins and the transcription machinery affect the transcription initiation and mRNA abundance of the target gene (Corton, Lapinskas and Gonzalez 2000).






DNA binding domain

Hinge region



Contains large binding pocket

Binds to variety of structure unrelated ligands

C terminal part of this domain

Agonist lock the receptor into an active conformation while antagonist stabilize the receptor into an inactive conformation

Transcriptionally active in the absence of ligands

Covalently modification by phosphorylation & sumuylation changes the ligand-binging affinity of the receptor

Ligand-binding domain (AF 2)

Ligand-independent transactivation domain (AF 1)

Ligand-binding domain (AF 2)

Permits the rotation of DNA-binding domain

Connects AB domain with EF domain

Most conserved domain

PPARs are targeted to a specific sequence (PPRE) of nucleotides within the regulatory regions of the responsive genes


Figure X: Structure of a PPARG protein proposed by Guo & Tabrizchi (2006)

Modified diagram from: Guo L., Tabrizchi R. (2006) Pharmacology & Therapeutics 111: 145-173

1.1.4. Function of the PPARG gene

The gene coding for PPARG lies on the short arm of chromosome 3 (3p25), spans about 146 kb of DNA and contains 7 exons. The PPARG gene was initially cloned from bone marrow cDNA library in 1995 (Greene et al. 1995). The principal site of expression of PPARG is the adipose tissues (Youssef and Badr 2004) although it is also expressed at the lower level in many other tissues such as the large intestine, colon, spleen, immune cells, endothelial cells and smooth muscle cells (Moraes, Piqueras and Bishop-Bailey 2006).

The reasons for selecting PPARG as one of our candidate genes is because it is functionally regulated by the dehydration product of prostaglandin D2 (PGD2), 15-deoxy- prostaglandin (15-dPGJ2), which is a major endogenous ligand of PPARG and associated with arachidonate-phospholipid metabolism (Na and Surh 2003). PPARG is also a modulator of insulin sensitivity, glucose homeostasis and blood pressure. Thiazolidinediones, a group of anti-diabetic drugs that improve insulin-resistance, are the artificial ligands of PPARG (Fajas et al. 1997). Also, PPARG is an important regulator of adipocyte differentiation and function.

Several genetic variations have been described in the human PPARG gene. The P115Q conversion was first identified in 4 morbidly obese subjects that cause defective phosphorylation of the protein resulting in accelerated adipocyte differentiation (Ristow, 1998). Two other mutations (P467L and V290M) were found in 3 individuals (in 2 families) with severe insulin resistance, dyslipidaemia and hypertension. Both of these mutations have been shown to cause impaired ligand-dependant transactivation and inhibit wild-type PPARG in a dominant negative manner (Barroso, 1999). Two other more common polymorphisms, a silent C to T substitution in exon 6 and a C (proline) to G (alanine) substitution at codon 12, have also been reported in PPARG (Vigouroux, 1998; Yen, 1997). There is evidence that suggests that the mutant protein produced by the alanine allele at codon 12 has reduced transcriptional and adipogenic, activity in vitro, which could lead to lower adipose tissue mass (Masugi, 2000). In addition, the alanine allele was found to be associated with lower lipolysis and greater insulin sensitivity in a group of lean non-diabetic subjects (Stumvoll, 2001).

1.1.4. Proposed genetic network

As both schizophrenia and type-2 diabetes are highly heritable diseases, they may share the same genetic network in some cases. Candidate gene studies and genome-wide association analyses have revealed dozens of genes associated with type-2 diabetes (E. Zeggini, L.J. Scott, et al, 2008). Of these associated genes, the peroxisome proliferator activated receptor-gamma (PPARG) locus might represent a pathological link between schizophrenia and type-2 diabetes as PPARG is functionally regulated by the dehydration product of prostaglandin D2 (PGD2), 15-deoxy-D prostaglandin (15-dPGJ2), which is a major endogenous ligand of PPAR and associated with arachidonate-phospholipid metabolism (H.K. Na and Y.-J. Surh, 2006).

Niacin-induced dermal flush has been used as a non-invasive diagnostic approach to test for deficiency in arachidonic acid (AA) released from membrane phospholipids and synthesis of AA derivatives such as prostaglandins (P.E. Ward, J. Sutherland, et al, 1998). Niacin is a form of vitamin B that functionally enhances the synthesis of PGD2, a prostaglandin that can stimulate blood vessels to dilate (J.D. Morrow, J.A. Award et al, 1992). Up to 80% of patients with schizophrenia have a poor skin response to niacin stimulation as compared to 20% of healthy individuals (P.E. Ward, J. Sutherland, 1998; B.K. Puri, T. Easton, 2001). These findings suggest that the diminished niacin flush may represent a subgroup of patients with schizophrenia who have abnormal phospholipid metabolism in their cellular membranes. If this were the case, such a subgroup might also have an increased risk for type-2 diabetes due to deficiency of PGD2 and 15-dPGJ2.

Phospholipase A2 (PLA2) is a group of the enzymes regulating the release of polyunsaturated fatty acids (PUFAs), such as AA, from the sn-2 position of phospholipids. This is an important step in the generation of PPARG ligands from arachidonate-phospholipids. The enzymatic activity of various members of the PLA2 family has been found to be altered in schizophrenia with some inconsistency (Gattaz et al 1987; Gattaz et al 1990; Gattaz et al. 1995; Albers et al. 1993; Noponen et al 1995; Ross et al. 1997; Ross et al. 1999; Smesny et al. 2005). Inhibitors of brain PLA2 activity have neuropharmacological effects and may prove to be therapeutically important in treating neuropsychiatric disorders (Farooqui et al. 2006). Genetic studies suggest that of the various PLA2 members, polymorphisms in the gene coding for PLA2 group 4 homolog A (PLA2G4A), a cytosolic form of PLA2, may contribute to the pathogenesis of schizophrenia (Law et al. 2006). Due to its role in prostaglandin generation and previous association with schizophrenia, PLA2G4A was selected for genotyping in this study.

Cyclooxygenase 2 (COX2 or PTGS2) is a rate-limiting enzyme responsible for prostaglandin synthesis in response to stimulus including some signaling pathways in the central nervous system (CNS) (Langenbach et al. 1999; Samad et al. 2001). Interestingly the genes coding for PTGS2 and PLA2G4A lie close to each other on the long arm of chromosomes 1. While PTGS2 was not associated with schizophrenia in a previous study using a smaller subset of this study population (Wei et al. 2004), it is worth genotyping in this larger population to see if there is a cis-phase interaction between these 2 loci. The present study therefore set out to test the hypothesis that the PLA2G4A, PTGS2 and PPAR genes are associated with schizophrenia singly or in a combined manner.

Hypothesis 1.2.: The role of the AKT1-GSK3 signalling pathway

AKT1 is another candidate gene we chose to study as it plays a functional role in intracellular signalling pathways that may be important for the pathophysiology of schizophrenia. Emamian and co-workers (2004) were the first to report on the AKT1 association with schizophrenia in a Northern European population. This initial finding has been replicated in a number of independent studies [Schwab et al., 2005; Bajestan et al., 2006; Ikeda et al, 2006; Norton et al., 2007; Xu et al., 2007; Thiselton et al., 2008], although some others failed to replicate the AKT1 association [Ohtsuki et al., 2004; Ide et al., 2006; Liu et al., 2006; Pinheiro et al., 2007; Turunen et al., 2007; Sanders et al., 2008]. The association of AKT1 and schizophrenia was also seen in other populations, Japanese, Irish, Iranian and Finnish.

1.2.1 Introduction

V-akt murine thymoma viral oncogene homologues (AKTs) are a subfamily of the phosphoinositide-dependent serine-threonine protein kinases and play a vital role in cellular processes such as glucose metabolism, cell proliferation, apoptosis, transcription and cell migration (Brazil and Hemmings 2001). AKTs also interact with neuronal dopaminergic signalling (Tan et al. 2008). There are three isoforms of AKT enzymes identified in mammalian cells, namely AKT1, AKT2 and AKT3, which are encoded by distinct genes located on chromosomes 14q32 (AKT1), 19q13 (AKT2) and 1q43 (AKT3) in humans. All three genes have greater than 85% identical DNA sequences and their protein products share the same structural organization (Kandel and Hay 1999). All 3 AKT isoforms are ubiquitously expressed in mammals although the levels of expression vary among tissues (Bellacosa et al. 1993). The most common isoforms prsent in most tissues is AKT1, which is highly expressed in regenerating neurons (Owada et al. 1997). The highest expression of AKT2 is observed in insulin-responsive tissues, such as skeletal muscle, heart, liver and kidney (Altomare et al. 1995), suggesting that this isoform is important for the insulin signaling pathway. Unlike AKT1 and AKT2, the AKT3 isoform shows a more restricted pattern of expression and higher levels are found in the testes and brain than in the adult pancreas, heart and kidney (Brodbeck, Cron and Hemmings 199).

1.2.2 Structure of AKT1

Figure X: Structure of the AKT molecule. (Walker et al. 1998)Walker et al in 1998 proposed that all three AKT-isoforms share similar structure. The detailed structure of AKT1 is shown in Figure V-1 and contains the following functional domains: (1) pleckstrin homology (PH) domain that binds phospholipids, (2) a catalytic domain, (3) a regulatory domain and (4) short glycine-rich region (bridges to the catalytic region). All three isoforms also possesse a conserved threonine and serine residue (threonine 308 and serine 473 in AKT1) and the phosphorylation of those sites is critical for full activation of AKT. The distance between the two phosphorylation sites/residues is around 160 - 170 amino acids (Kandel and Hay, 1999).

1.2.3 The proposed pathway

AKT molecule is activated inside the cell. Briefly, generation of D3-phosphorylated phosphoinositides by phosphoinositide 3-kinases (PI3K) induce the translocation of AKT on the plasma membrane via its PH domain where it is activated. In the plasma membrane, AKT co-localizes with the constitutively active phosphoinositide-dependent kinase-1 (PDK-1), which activates AKT by phosphorylation on the different residues, for example, Thr308 and Ser473 in the AKT1 isoform (Kandel and Hay, 1999). After activation, AKT can phosphorylate a number of substrates, one of which is glycogen synthase kinase 3 (GSK-3) (Norton et al. 2007). On phosphorylation, AKT1 inhibits GSK-3 that upregulates the glycogen synthase activity, which is part of the insulin signal transduction pathway (Ali, Hoeflich and Woodgett 2001). Impaired AKT1-GSK3 beta signaling has been reported in schizophrenia (Emamian et al. 2004). Moreover, GSK3 has also been implicated in the multifactorial etiology of skeletal muscle insulin resistance in animal models and in patients with type 2 diabetes (Dokken, Sloniger and Henriksen 2005). The dysfunction of the AKT1-GSK3 signal pathway may be involved in both type-2 diabetes and schizophrenia. Several studies have shown the ATK1 association with schizophrenia (Emamian et al. 2004; Neuton et al., 2007).

The present study therefore was to replicate the initial findings in the British samples and also to explore if the AKT1 gene was involved in development of antipsychotics induced type-2 diabetes.

Effects of clozapine on expression of the genes associated with type-2 diabetes and obesity

The patients who suffer from schizophrenia have to be on antipsychotic medication. The major side effect of these drugs is tremendous weight gain, hence obesity. The major risk factor of type-2 diabetes is obesity. Therefore we tested the effect of clozapine, an atypical antipsychotic drug, on the expression of genes associated with obesity and type-2 diabetes identified by genome-wide association studies through tissue culture techniques. These studies have recently identified more than 20 genes for obesity and type-2 diabetes (Table X).We also tested the candidate genes selected in the above mentioned hypothesis and also the cytokine genes.