- Research article
- Open Access
- Open Peer Review
Early peroxisome proliferator-activated receptor gamma regulated genes involved in expansion of pancreatic beta cell mass
© Vivas et al; licensee BioMed Central Ltd. 2011
- Received: 28 July 2011
- Accepted: 30 December 2011
- Published: 30 December 2011
The progression towards type 2 diabetes depends on the allostatic response of pancreatic beta cells to synthesise and secrete enough insulin to compensate for insulin resistance. The endocrine pancreas is a plastic tissue able to expand or regress in response to the requirements imposed by physiological and pathophysiological states associated to insulin resistance such as pregnancy, obesity or ageing, but the mechanisms mediating beta cell mass expansion in these scenarios are not well defined. We have recently shown that ob/ob mice with genetic ablation of PPARγ2, a mouse model known as the POKO mouse failed to expand its beta cell mass. This phenotype contrasted with the appropriate expansion of the beta cell mass observed in their obese littermate ob/ob mice. Thus, comparison of these models islets particularly at early ages could provide some new insights on early PPARγ dependent transcriptional responses involved in the process of beta cell mass expansion
Here we have investigated PPARγ dependent transcriptional responses occurring during the early stages of beta cell adaptation to insulin resistance in wild type, ob/ob, PPARγ2 KO and POKO mice. We have identified genes known to regulate both the rate of proliferation and the survival signals of beta cells. Moreover we have also identified new pathways induced in ob/ob islets that remained unchanged in POKO islets, suggesting an important role for PPARγ in maintenance/activation of mechanisms essential for the continued function of the beta cell.
Our data suggest that the expansion of beta cell mass observed in ob/ob islets is associated with the activation of an immune response that fails to occur in POKO islets. We have also indentified other PPARγ dependent differentially regulated pathways including cholesterol biosynthesis, apoptosis through TGF-β signaling and decreased oxidative phosphorylation.
- Insulin Resistance
- Gene Ontology
- Beta Cell
- Ingenuity Pathway Analysis
- KEGG Pathway
Although the hallmark of obesity associated type 2 diabetes (T2D) is the decrease in insulin sensitivity, the development of hyperglycemia requires the failure of the allostatic response of the β-cells to respond by producing enough insulin to overcome the functional defect in insulin action . One of the strategies the endocrine pancreas uses to adapt to changes in insulin resistant requirements associated with different physiological states, such as pregnancy, obesity, or ageing, is to expand the β-cell mass. Thus, in all these states insulin resistance leads to an increased production of insulin to maintain euglycemia . Despite the increased requirements, the majority of individuals remain euglycemic by adequately increasing their β-cell mass and by adjusting their stimulated insulin secretion. However, when the allostatic β-cell adaptation fails, hyperglycemia will develop. Under conditions of allostatic overload, there is an association between pregnancy with gestational diabetes and obesity and ageing with T2D .
In humans and animal models, it has been widely recognised that β-cell failure is an essential factor leading to T2D. This can occur when β-cells fail to appropriately expand and/or to optimise their function, ultimately compromising in glucose-stimulated insulin secretion (GSIS). Animal models of insulin resistance are excellent models to demonstrate the plasticity of β-cell mass and provide suitable experimental systems in which to identify the extracellular signals and molecular mechanisms behind this compensatory response.
Valuable insights into the key role of β-cell failure in the pathogenesis of T2D has come from genome-wide association studies, an important resource to identify new unexpected susceptibility gene candidates for the development of T2D . Of interest these studies identified validated variants associated with insulin-secretory defects in the general population and showed little if any relationship to insulin resistance [5–9].
Peroxisome proliferator-activated receptor gamma (PPARγ) is a member of the nuclear receptor superfamily of ligand-activated transcription factors  and has been shown to be involved in many diverse biological processes, including adipogenesis and glucose and lipid metabolism. In addition to these roles, it has been shown that PPARγ also exerts an important role in controlling cellular proliferation in different organs including pancreatic endocrine tissue. As already demonstrated, the absence of PPARγ specifically in β-cells cannot fully compensate for the β-cell dysfunction seen in states of peripheral insulin resistance . In fact, animals with lack of PPARγ had blunted the physiological expansion of β-cell mass in response to high-fat feeding.
There are important differences between the mechanisms controlling the increase of β-cell mass observed during pregnancy and obesity suggesting certain degree of etiopathogenic specificity on the mechanisms controlling β-cell mass expansion [12–15]. We have recently shown that β-cell mass adaptation to insulin resistance failed in the adult POKO mice, an insulin resistant mouse resulting from the deletion of PPARγ2 in an obese ob/ob background . This impaired β-cell mass expansion contrasted with the massive expansion observed in their obese littermate ob/ob mice. Thus, we rationalised that comparison of ob/ob and POKO mice islets particularly at early ages could provide some new insights on early PPARγ dependent transcriptional responses involved in the process of β-cell mass expansion.
To date, the main approach to identifying global transcriptional changes has been the use of gene expression microarray platforms in advanced stages of the disease. Here we focused in early evolutive stages to increase our chances of identifying novel early events and their molecular effectors involved in the adaptation of β-cells to insulin resistance. Our strategy has been to isolate total RNA samples from islets of 5-week-old female wild type (WT), PPARγ2KO, ob/ob and POKO mice for large-scale expression profiling. We have used 5-week-old mice to elucidate pathways and factors underlying the early islet proliferative response that might have failed in the POKO mice and that could cause the inadequate β-cell expansion in this model. Although at this age metabolic disturbances have not been identified in islets yet, insulin resistance is already detectable in ob/ob and POKO mice. In support of this approach we have shown that defects in proliferation markers can be detected at 5 weeks of age in POKO islets, well before alterations in the insulin secretory capacity were evident .
Initially, we have analyzed genes to compare ob/ob with WT mice. This comparison should identify any gene regulation in ob/ob mice that may be linked to the increased expansion of their β-cells. Secondly, we have compared POKO vs. WT islets. As these two models have a similar degree of expansion of their β-cells, the aim was to identify default genes induced during insulin resistance that were dependent on PPARγ2 activity. In our previous study  we have shown that at 16 weeks of age, despite being more insulin resistant than ob/ob mice, the β-cell mass of POKO mice remained similar to that of WT islets. This resulted in POKO mice having an inappropriately lower insulin plasma level than the ob/ob controls at 16 weeks of age . The final comparison that we have analyzed was between POKO and ob/ob islets, representing different states of insulin resistance that could identify PPARγ-dependent genes involved in expansion and function of β-cell.
In this study we have identified genes that are downregulated in PPARγ-deficient POKO islets that are related to the proliferation and survival mechanisms that may facilitate the expansion of β-cell mass under insulin resistant conditions. We have also identified new pathways that may contribute to this adaptation such as immune response, cholesterol biosynthesis, apoptosis through TGF-β signaling pathway and decreased oxidative phosphorylation. Failure of these adaptive mechanisms may contribute to failure of β-cell function in POKO islets and may be dependent on PPARγ expression and functionality.
Differential expressed genes amongst the specific comparisons
ob/ob vs. WT
PPARγ2KO vs WT
POKO vs. WT
ob/ob vs. PPARγ2KO
POKO vs. ob/ob
POKO vs. PPARγ2KO
7 (p ≤ 0.2)
1 (p ≤ 0.3)
81 (p ≤ 0.2)
26 (p ≤ 0.2)
3 (p ≤ 0.3)
85 (p ≤ 0.2)
Differential gene expression between wild type and ob /ob mice islets
Differential gene expression between WT and ob/ob islets and between WT and POKO islets
Comparison WT vs. ob/ob
cyclin A2 (Ccna2), mRNA
cyclin B1 (Ccnb1), mRNA
Cyl-1 = cyclin D1 [mice, BALB/c, brain, mRNA, 3737 nt].
cyclin D3 (Ccnd3), transcript variant 1, mRNA
F-box and leucine-rich repeat protein 21 (Fbxl21), mRNA
F-box protein 27 (Fbxo27), mRNA
Mus musculus IL2-inducible T-cellkinase (Itk), mRNA
tumor necrosis factor receptor superfamily, member 11b (osteoprotegerin) (Tnfrsf11b), mRNA
baculoviral IAP repeat-containing 5 (Birc5), transcript variant 1, mRNA
F-box and WD-40 domain protein 14 (Fbxw14), mRNA [NM_015793]
cyclin-dependent kinase 6 (Cdk6), mRNA
blastocyst cDNA, RIKEN full-length enriched library, clone:I1C0038J11 product:cyclin-dependent kinase 7 (homolog of Xenopus MO15 cdk-activating kinase), full insert sequence
Comparison WT vs. POKO
tumor necrosis factor receptor superfamily,member 11b (osteoprotegerin)(Tnfrsf11b), mRNA
mRNA for Ki-67.
surfactant associated protein D (Sftpd), mRNA
free fatty acid receptor 2 (Ffar2), mRNA
As expected the comparison between islets from ob/ob and WT mice identified many of the genes previously identified that play a role in β-cell expansion during states of insulin resistance such as obesity and pregnancy . These positive controls included genes involved in proliferation, cell cycle and survival mechanisms . For instance, gene expression levels of cyclin A2, B1, D1 and D3 were significantly increased in islets from ob/ob compared with WT mice. Conversely expression of cell division protein kinase 6 (Cdk6) and Cdk7 were significantly decreased in the ob/ob compared to WT islets. This transcriptional pattern closely mirrors the pattern observed during the β-cell hypertrophy typically associated with pregnancy, suggesting these genes are part of a common proliferative response associated with modulation of β-cell mass in physiological and pathophysiological states. Our analysis also confirmed the upregulation of genes encoding for enzymes involved in regulation of β-cell expansion (IL2-inducible T-cell kinase (Itk), Tumor necrosis factor receptor superfamily, member 11b (osteoprotegerin or Tnfrsf11b), baculoviral IAP repeat containing 5 (Birc5)) and cell cycle regulation (F-box and leucine-rich repeat protein 14 (Fbxw14), Fbxl21 and F-box protein 27 (Fbxo27)). Globally, these data indicate that at this early age the ob/ob islets have already activated their proliferative machinery and this data can be considered as a positive quality control for the experiment.
Differential gene expression between WT and POKO islets
We next compared the gene expression profile between POKO and WT islets (Table 2). Since the POKO mouse fails to expand its β-cell mass despite insulin resistance, we hypothesized that this list may reveal by default genes induced during insulin resistance that were dependent on PPARγ2 activity. Our first observation was that POKO and WT islets had a very similar gene expression profile. In fact the expression of many of the genes previously shown to be dysregulated in ob/ob islets were not significantly different in the analysis comparing POKO and WT islets. This is consistent with the maintenance of "normal" β-cell mass in POKO mice that fail to expand. However, through more detailed analysis, it revealed that despite the inability to expand their β-cell mass, POKO islets had induction of some of recognized mediators of β-cell expansion, as observed during both pregnancy and obesity. These included Tnfrsf11b, mKi67, Surfactant-associated protein D (Sftpd) and Free fatty acid receptor 2 (Ffar2). These genes were increased in both the comparison between ob/ob vs. WT, and also to a lesser extent in the comparison between POKO vs. WT. These data suggested a dysregulation of the homeostatic control of β-cell mass in POKO islets, and that the ability of POKO mice to appropriately induce an appropriate response to a metabolic stress is impaired, with no increased proliferation and activation of survival mechanisms in their islets. Globally considered, this indicates a partial response of PPARγ-deficient ob/ob islets to insulin resistance.
Differential gene expression between ob/ob and POKO mice
We hypothesized that the POKO vs. ob/ob comparison could identify PPARγ2 islet target genes that were dysregulated during the early stages of insulin resistant states contributing to a lack of β-cell mass expansion. We have previously shown that at 16 weeks of age POKO mice have significantly decreased islet number and mass compared to ob/ob mice and more interestingly, the number and size of islets in POKO mice resembled the number and size of the islets of a wild type 16-week-old mouse [17, 18]. Analysis of gene expression by RT-PCR showed that at 5 weeks of age, there were already some changes in the expression of specific genes relating to proliferation (CyclinD1) and β-cell function (Insulin, MafA, PDX1 and Glut 2), with no evident differences in insulin secretory defects between POKO and ob/ob . However, differences in gene expression of other genes, which have been shown at 16 weeks were inconsistent at this age. Surprisingly the comparison between 5-week-old ob/ob and POKO islets on microarray data only identified 3 upregulated genes, (Actin alpha1 (Acta1), Thrombospondin 4 (Thbs4) and Solute carrier family 5 (sodium/glucose cotransporter) (Slc5a10), and as expected, down regulation of gene PPARγ with an adjusted p-value ≤0.3 (Table 1). To allow a more detailed comparison between our models, we performed pathway analysis using Fatigo and FatiScan tools, enabling significant multiple comparisons of genes in significant pathways despite not having differentially specific expressed genes from these pathways at this early age.
Dysregulated gene expression pathway analysis in models with different degree of beta cell expansion
A subsequent post-analysis was performed on the microarray data from the comparison between 5-week-old islets in order to identify the main dysregulated pathways, relating to the expansion of β-cells between the three genotypes. For this we have used two bioinformatic tools: the Ingenuity Pathway Analysis (IPA) and gene enrichment test from Babelomics tools (Fatigo), using an adjusted p value ≤0.05. To allow reliable interpretation of our data we have made the following comparisons: ob/ob vs. WT; associated with the normal adaptation to obesity induced insulin resistance with an expansion of the β-cell; and POKO vs. WT; associated with a failure in the adaptive response, with no expansion of β-cell. Using these two lists, we analyzed up-regulated and down-regulated genes from the comparison between ob/ob and WT, which were not regulated between POKO and WT.
Functions and canonical pathways as differentially expressed up-regulated genes in ob/ob vs. WT, with no significant changes in POKO vs. WT mice
Top Bio Functions
Diseases and Disorders
7.66E-18 - 3.63E-03
5.95E-15 - 3.60E-03
2.28E-14 - 3.63E-03
Skeletal and Muscular Disorders
4.99E-12 - 3.63E-03
Connective Tissue Disorders
7.78E-12 - 2.44E-03
Molecular and Cellular Functions
1.45E-17 - 4.37E-03
Cellular Function and Maintenance
2.21E-15 - 3.58E-03
Cell-To-Cell Signaling and Interaction
9.41E-15 - 4.37E-03
Cellular Growth and Proliferation
3.93E-13 - 3.11E-03
1.94E-09 - 3.78E-03
Physiological System Development and Function
Hematological System Development and Function
3.63E-16 - 4.04E-03
3.63E-16 - 3.27E-03
2.59E-15 - 3.60E-03
Cell-mediated Immune Response
1.22E-14 - 3.78E-03
Immune Cell Trafficking
2.25E-13 - 3.78E-03
Top Canonical Pathways
Autoimmune Thyroid Disease Signaling
Allograft Rejection Signaling
Type I Diabetes Mellitus Signaling
Graft-versus-Host Disease Signaling
B Cell Development
Functions and canonical pathways as differentially expressed down-regulated genes in ob/ob vs. WT, with no significant changes in POKO vs. WT mice
Top Bio Functions
Diseases and Disorders
Dermatological Diseases and Conditions
3.01E-05 - 4.21E-02
3.01E-05 - 4.21E-02
9.80E-05 - 4.21E-02
6.95E-04 - 4.21E-02
1.48E-03 - 4.21E-02
Molecular and Cellular Functions
4.46E-07 - 4.21E-02
Small Molecule Biochemistry
4.46E-07 - 4.21E-02
Vitamin and Mineral Metabolism
4.46E-07 - 4.21E-02
Cell-To-Cell Signaling and Interaction
1.48E-04 - 4.21E-02
Cellular Assembly and Organization
1.48E-04 - 4.21E-02
Physiological System Development and Function
9.80E-05 - 4.21E-02
Hair and Skin Development and Function
1.37E-03 - 4.21E-02
1.37E-03 - 4.18E-02
Digestive System Development and Function
1.50E-03 - 1.95E-02
1.50E-03 - 4.18E-02
Top Canonical Pathways
Biosynthesis of Steroids
HER-2 Signaling in Breast Cancer
Tight Junction Signaling
Fatty Acid Biosynthesis
KEGG significant terms up-regulated in ob/ob vs. WT, without significant changes in POKO vs. WT mice
Odds ratio log
Type 1 diabetes
Graft-verus host disease
Autoinmune thyroid disease
Cell adhesion molecules
Long term depresion
Vascular smooth muscle contact
InterPro significant terms up-regulated in ob/ob vs. WT, without significant changes in POKO vs. WT mice
Odds ratio log
2-5-oligoadenylate synthetase, ubiquitin-like region (IPR006116)
Regulator of G protein signalling (IPR000342)
Immunoglobulin C1-set (IPR003597)
Protein of unknown function DUF622 (IPR006907)
Immunoglobulin V-set, subgroup (IPR003596)
Immunoglobulin/major histocompatibility complex, conserved site (IPR003006)
Immunoglobulin V-set (IPR013106)
Immunoglobulin subtype (IPR003599)
Immunoglobulin subtype 2 (IPR003598)
GO significant terms up-regulated in ob/ob vs. WT, without significant changes in POKO vs. WT mice
Odds ratio log
GO molecular function
antigen binding (GO:0003823)
GO cellular component
cell surface (GO:0009986)
external side of plasma membrane (GO:0009897)
guanylate cyclase complex, soluble (GO:0008074)
GO biological process
immune response (GO:0006955)
peptide hormone processing (GO:0016486)
GO significant terms down-regulated in ob/ob vs. WT, without significant changes in POKO vs. WT mice
Odds ratio log
GO molecular function
L-glutamine transmembrane transporter activity (GO:0015186)
GO cellular component
apical junction complex (GO:0043296)
apicolateral plasma membrane (GO:0016327)
tight junction (GO:0005923)
cell-cell junction (GO:0005911)
cell junction (GO:0030054)
GO biological process
cholesterol biosynthetic process (GO:0006695)
sterol biosynthetic process (GO:0016126)
tongue development (GO:0043586)
glutamine transport (GO:0006868)
Functional profiling of differentially expressed genes in POKO mice that may contribute towards failure to expand beta cell mass
We performed FatiScan analysis in the comparison between POKO and ob/ob islets obtained from our microarray. This FatiScan analysis highlights the value of using functional analysis, which incorporates all experimental data rather than limiting interpretation to those genes that rank among the highly differentially expressed. Using the FatiScan tool, 616 gene ontology groups, 85 InterPro, 67 KEGG pathways, 7 Biocarta, and 3 JasparTBS were observed to be significantly (adj. p value ≤ 0.05) overrepresented in islets from POKO relative to the ob/ob mice. There was a substantial overlap of genes within these identified functional groups resulting in the overrepresentation of a large number of functionally similar gene ontology groups. Therefore only the highly expressed groups have been shown in tables.
Initially, an overrepresentation of genes related to biological processes such as neutral lipid catabolic process, lipid glycosilation, collagen biosynthesis process and regulation of actin filament length were identified. Analysis from FatiScan also identified genes related to mitochondrial outer and inner membrane translocase complex and fibrillar collagen in the cellular component group and profilin, lipoic acid and IgG binding in the molecular function group. Other molecular functions associated with highly expressed genes in POKO islets compared with ob/ob were RPTP-like protein binding, platelet-derived growth factor binding and nucleocytoplasmic transporter activity. Genes from terms associated to Chaperonin TCP-1, Mob/phocein, hexokinase were identified as domains in InterPro.
KEGG significant terms up-regulated in POKO vs. ob/ob mice
Regulation of actin cytoskeleton
Pathways in cancer
Leukocyte transendothelial migration
p53 signaling pathway
Calcium signaling pathway
Cysteine and methionine metabolism
Glycine, serine and threonine metabolism
Terpenoid backbone biosynthesis
ErbB signaling pathway
Nucleotide excision repair
NOD-like receptor signaling pathway
Arginine and proline metabolism
MAPK signaling pathway
Chemokine signaling pathway
TGF-beta signaling pathway
Alanine, aspartate and glutamate metabolism
Cytokine-cytokine receptor interaction
Amino sugar and nucleotide sugar metabolism
Valine, leucine and isoleucine biosynthesis
Pentose and glucuronate interconversions
PPAR signaling pathway
KEGG significant down-regulated in POKO vs. ob/ob mice
Vascular smooth muscle contraction
Type I diabetes mellitus
Antigen processing and presentation
Cardiac muscle contraction
Phosphatidylinositol signaling system
Autoimmune thyroid disease
GnRH signaling pathway
Type II diabetes mellitus
Progesterone-mediated oocyte maturation
Maturity onset diabetes of the young
Carbohydrate responsive element-binding protein (Chrebp), casein kinase 1 (ck1) and insulin-like growth factor type 1 receptor (igfr1) pathways were the three highly expressed in Biocarta associated to genes with higher expression in POKO compared with ob/ob islets (data not shown).
Jaspar TFBS significant terms down-regulated in POKO vs. ob/ob mice
Odds ratio log
Dysregulation of genes dependent on PPAR in POKO and ob/ob islets
One of the pathways we found to be altered in FatiScan analysis between POKO vs. ob/ob was the PPAR pathway (Table 9). The expression of PPAR alpha and total PPAR gamma were decreased in the islets of POKO compared with ob/ob mice (Figure 2); however the expression of PPAR beta/delta were increased (Figure 2 and Additional file 1 Table S1). Furthermore, the expression of the three isoforms of the nuclear receptor Retinoic X receptor (RXR alpha, beta and gamma), heterodimers of PPARs, were differentially altered depending on the isoform as shown in the table. Most of the pathways analysed had increased or decreased expression of PPAR target genes as shown in Additional file 1 Table S1. However, gene expression of PPAR target genes involved in lipogenesis (SCD1, ME1, FADS2) and in cholesterol metabolism (CYP7A1, CYP8B1, LXRa and CYP27) were increased in POKO compared to ob/ob islets (Figure 2). Conversely, PPAR target genes involved in fatty acid oxidation (Ehhadh, Acaa1b, SCP2, Acox1, CPT1a, CPT1c and CPT2) were decreased suggesting these genes could be regulated by the PPARγ2 isoform, a pattern also observed with genes involved in adipocyte differentiation (Perilipin, aP2, adipoq, MMP-1) or gluconeogenesis (PEPCK, GyK, AQP7).
Genome Wide candidate genes and early transcriptional response in pancreatic beta cells
Recent genome-wide association studies have provided an important resource for furthering our understanding of type 2 diabetes disease mechanisms [5, 9]. These published variants were all associated with insulin-secretory defects in the general population and show little if any relationship to insulin resistance. We have checked these genes in both our comparisons (See additional file 1 Figure S3). Although expression of Slc30A8 was not altered in the two comparisons in our microarray data, we found that expression of Slc30A8 detected by RT-PCR was decreased in islets from POKO compared to ob/ob mice and in islets from PPARγ2KO compared to WT mice. Furthermore, using NSITE/Recognition of Regulatory motifs tool we found a PPAR conserved sequence motif in this gene suggesting this gene has a dependency on PPARγ.
We also checked expression of NR4A3 (also called Neuron-derived orphan receptor (Nor) 1) by RT-PCR in our models. This is a novel candidate gene for β-cell function, which was not covered by the SNP arrays of the recent genome-wide association studies for type 2 diabetes mellitus . Here we found significantly increased expression of NR4A3 in ob/ob and POKO islets when compared with WT and PPARγ2KO islets, but no significant differences between the ob/ob and POKO islets (See Additional file 1 Figure S3).
Recent studies have shown that the Notch-regulated transcription factor neurogenin 3 (Ngn3) is critical for the development of endocrine cells of islets . Ngn3 was increased in islets from ob/ob and POKO mice compared with those from PPARγ2KO and WT, but with no significant differences between the former genotypes. From the current GWAS and metaanalysis in Europeans associated to adiposity loci , we identified increased NEGR1 expression in our microarray data between ob/ob and WT islets and lower expression in POKO when compared to ob/ob islets after analysis by RT-PCR.
In this study we have used a gene expression microarray and a subsequent bioinformatic approach to identify new molecular mechanisms and pathways implicated in the expansion of pancreatic β-cells in the context of insulin resistance. Using this technique and bioinformatic analysis we have investigated: 1) the overrepresentation of genes related to β-cell expansion under situations of insulin resistance in islets, 2) the genes differentially expressed in insulin resistant islets with and without the capacity of β-cell to expand, and 3) the overrepresentation of functional groups of genes differentially expressed previously to normal o defective expansion of β-cell mass.
We first examined genes that were upregulated in the islets of ob/ob compared to WT mice, expecting to identify pathways relevant to β-cell mass expansion for future comparison with the POKO mouse which is unable to expand its islets. We found that, at the age of 5 weeks, the ob/ob islets already exhibit induction of several pathways relating to cell proliferation and survival mechanisms. This response was not observed in POKO islets suggesting that these regulatory mechanisms are activated very early on before severe metabolic stress has developed, and that a failure to initiate these pathways is effectively contributing to the failure of the homeostatic control of β-cell mass in POKO islets. Specifically, POKO islets did not show the same induction of cyclin A2, B1, D1 and D3 observed in ob/ob islets, where in addition to their role in cell cycle progression, new data has revealed an emerging role of D-type cyclins in transcriptional regulation and cellular differentiation processes. There is also evidence from studies in 3T3-L1 cells that cyclin D3 acts as a ligand-dependent PPARγ coactivator and that its expression is increased throughout the adipocyte differentiation process . Our results could therefore suggest that regulation of cyclinD3 is at least partly dependent on PPARγ which may be of relevance in the proliferative mechanisms of β-cell mass. Other universal genes previously shown to be increased during the expansion of the β-cell during obesity or pregnancy such as Birc5, Igf1r and Rasgrp1 also failed to increase in PPARγ deficient ob/ob islets (POKO vs. WT) further suggesting an inability of POKO islets to proliferate at a young age.
We were interested in identifying a putative molecular mechanism that could induce the pancreatic islet dysfunction in POKO mice. Our experimental design allowed the identification of pathways associated with the normal adaptation of islets to obesity induced insulin resistance (comparison ob/ob vs. WT) and pathways associated with failure in this response (comparison POKO vs. WT). We have identified genes associated with canonical pathways of Autoimmune Thyroid Disease Signalling, Allograft Rejection Signalling and Type 1 Diabetes Mellitus Signalling are upregulated in ob/ob vs. WT islets. This was associated with a significant increase of genetic programmes related to inflammatory disease, inflammatory response and immunological disease. It is well established that inflammation and fibrosis are processes preceding the deterioration of pancreatic islet structure. Molecules from MHCII (HLA-D, HLA-A, HLA-B) and other molecules such as CD80/86, which are involved in the antigen presentation signalling were increased in ob/ob relative to WT islets, with no significant changes in the islets of POKO vs. WT, suggesting that their involvement in remodelling is linked to β-cell mass expansion. As these pathways are activated in ob/ob islets and not in POKO islets, and that this is occurring very early before ob/ob islets fail, we speculate that this immune and inflammatory response may in fact be a physiological response, initiating or enabling a mechanism of remodelling or adaptation in order for the β-cell mass to expand. Thus it is conceivable that these changes are part of an adaptive response required under metabolic stress as occurs in ob/ob mice, possibly involving signalling via PPARγ.
Our FatiScan analysis showed that TGF-β/Smad signaling pathway was significantly increased in POKO compared to ob/ob islets, which is supported by increased pancreatic staining of TGF-β in POKO islets at 4 and 16 weeks of age. It has already been shown in other organs such as kidney that stimulation of fibrotic processes could be mediated, at least in part, by a down-regulation of PPARγ that can favour an up-regulation of the TGF-β/Smad signalling pathway [24, 25]. Genes such as TGF-β1, its receptors type I and II and SMAD2/3/4, also implicated in apoptosis and cell cycle arrest through this pathway were increased in POKO islets compared to ob/ob islets, suggesting that specific profibrotic and apoptotic genes could also be involved in the failure of β-cell expansion in POKO islets.
Two genes, perforin 1 and granzyme B which are required for the direct recognition of the cells  were also up regulated in our ob/ob vs. WT comparison. The dominant role of the perforin/granzyme pathway in β-cell destruction in type 1 diabetes and allogeneic islet graft rejection makes this pathway an important target for future therapies for type 1 diabetes. However our results and the suggested new role of these molecules in inflammation also suggests that they could be new targets for treatments of type 2 diabetes, as part of their role includes the adaptation of the β-cell to insulin resistance. Indeed they could contribute to the remodelling mechanisms enabling the necessary expansion of the pancreatic β-cell by controlling apoptosis in a PPARγ dependent pathway. Supporting this idea, we have found that the perforin promoter has a PPAR response element and that POKO islets lack this immune response, further supporting its role facilitating the remodelling adaptation in response to insulin resistance. Globally considered our data indicate that PPARγ-ablated islets (POKO) may have decreased capacity for remodelling and an exacerbated proapoptotic response ultimately leading to their failure.
Besides an increased β-cell mass during the adaptation to insulin resistance, the most immediate response is the optimisation of the insulin secretion. The data from this study showed that cholesterol-induced impairment of β-cell function could occur in POKO islets as happens in other models . It has already been shown that Ldlr-/- knockout mice with β-cell-specific ATP-binding cassette transporter A1 (ABCA1) deficiency showed increased islet cholesterol content and β-cell dysfunction, suggesting that cholesterol efflux through ABCA1 is a critical regulator of islet cholesterol content and β-cell function . Functional analysis using FatiScan and our RT-PCR results revealed increased expression of genes related to cholesterol and sterol biosynthesis and decreased expression of ATP-binding cassette (ABC) transporters in POKO compared with ob/ob islets. Also rosiglitazone, an activator of the peroxisome proliferator-activated receptor-gamma, requires β-cell ABAC1 for its beneficial effects on glucose tolerance. It has been suggested that elevated β-cell cholesterol levels may impair insulin secretion by promoting the dimerisation of neuronal NO synthase (nNOS), which downregulates glucokinase (GK), thus impairing glucose sensing. This will agree with our previous observation showing that GK is downregulated in POKO compared with ob/ob islets . Moreover, it has been suggested that cholesterol also inhibits steps in insulin exocytosis and that ABCA1 could have other independent effects on insulin secretion. Here we show that granulophilin is downregulated in POKO islets and its expression depends on PPARγ similarly to GK. Thus we suggest that PPARγ could play a main role determining the cholesterol flux and its toxicity on the islet insulin secretion machinery under insulin resistance conditions.
Another pathway down regulated in POKO in comparison to ob/ob islets was oxidative phosphorylation. Different isoforms of Cytochrome c oxidase and NADH dehydrogenase were downregulated in POKO compared to ob/ob islets suggesting impaired complexes of the mitochondrial oxidative phosphorylation system in POKO islets could contribute to the failure of their islets. It has been shown recently in human islets and cell lines , that increased levels of nitrogen species may interact with the insulin exocitosis mechanism. From our previous data, we did not observe increased levels of ROS in POKO mice at this age; however nitrogen species under pathophysiological conditions such as chronic hyperglycemia could result in insulin secretion dysfunction in POKO islets.
Actin cytoskeleton remodelling is known to be involved in glucose-stimulated insulin secretion (GSIS). Cell-cell contacts mediated by intercellular junctions are crucial for proper insulin secretion in the endocrine pancreas. FatiScan analysis found that the tight junction, adherens junction and focal adhesion pathways were increased in POKO compared to ob/ob islets. In fact expression of Actin 1 alpha, which is involved in focal adhesion, was increased in POKO islets. Of note activation of actin cytoskeleton could result in increased fibrosis in the islets of POKO mice in a process facilitated by TGF-β/Smad activation as has been similarly shown in liver fibrosis .
Genome-wide association studies (GWAS) have had success in identifying loci that are involved in common diseases such us diabetes and interestingly some locus were linked specifically with T2D. Mouse models offer a good tool to help to take data from GWAS studies further investigating the function of associated genes in vivo. We further investigated some of these T2D susceptibility loci identified thorough GWAS in our models. We found only significant dowregulation in GK and SLC30A8 in POKO in comparison to ob/ob islets, both dependent on PPARγ expression. Specific β-cell inactivation of SLC30A8 in mouse results in glucose intolerance  and recently published meta-analysis results revealed a significant association between the rs13266634 C/T polymorphism of SLC30A8 and T2DM and IGT . Our results suggest that SLC30A8 might therefore have a role in expansion of the β-cell and benefit from PPARγ agonists in the treatment of type 2 diabetes.
Using transcriptome tools and mouse models of insulin resistance, in this study we show: a) that mechanisms of proliferation and survival signals were downregulated in POKO islets from an early age, b) we also identified new pathways that could be involved in the adaptation of the β-cell to insulin resistance during obesity and that they are defective in PPARγ-deficient ob/ob islets. Our results indicate that early stages of β-cell mass expansion are associated with activation of an inflammatory immune response that could facilitate the necessary remodeling while preventing activation of an apoptotic cascade. Other potential relevant pathogenic mechanisms modulating β-cell mass include cholesterol toxicity, TGF-β/Smad pathway-induced apoptosis/fibrosis and defective oxidative phosphorylation. As the pathways identified in our POKO islets are involved in the failure of β-cell function, our results indicate an important role of PPARγ in islets, modulating the adaptive response of β-cell mass to increased metabolic demands imposed by obesity and insulin resistance.
Animals were housed in a temperature-controlled room (20-22 8C) with 12-h light/dark cycles. Food and water were available ad libitum unless noted. All animal protocols used in this study were approved by the UK Home Office and the University of Cambridge and by Ethical and Veterinary committees of Rey Juan Carlos University in Spain.
Isolation and culture of pancreatic islets
The pancreas was injected, though the bile duct, with cold Hank's solution containing 0.4% (w/v) collagenase P (Roche Biochemicals). The pancreas was removed, digested for 8 minutes, and islets were collected by hand picking. Isolated islets were cultured overnight in cell medium (RPMI 1640 with 10% FBS, and 100 UI/ml penicillin, 100 μg/ml streptomycin), at 37°C, in 5% CO2 in air. For RNA extraction, islets were used on the day after isolation.
RNA isolation and purification
An average of 25-50 islets were used per sample and total RNA was extracted according to the manufacturer's instructions. Each sample was purified using the RNeasy® Mini kit (Qiagen Ltd, Crawley, UK). RNA was quantified using a NanoDrop® ND1000 spectrophotometer V 3.5.2 (NanoDrop Technologies, Wilmington, DE). RNA quality was subsequently assessed using the 28S:18S ratio running the samples on a denaturing agarose gel. The 2:1 ratio (28S:18S) indicated that the RNA was intact.
One-Color Microarray-Based Gene Expression Analysis Protocol (Agilent Technologies, Palo Alto, CA, USA) was used to amplify and label RNA. A pool of 3-4 mice were used per each microarray and 3 microarrays were used per each genotype. Briefly, 400 ng of total RNA from WT, PPARγ2KO, ob/ob and POKO mice, were reverse transcribed using T7 promoter primer and the Moloney murine leukemia virus (MMLV) reverse transcriptase. cDNA was then converted to anti-sense RNA (aRNA) by using T7 RNA polymerase that amplifies target material and incorporates cyanine 3 (Cy3)-labeled CTP simultaneously.
Sample hybridization and image analysis
Samples were hybridized to a Whole Mouse Genome Microarray 4x44K (G4122F, Agilent Technologies). 1.65 micrograms of Cy3-labeled aRNA were hybridized for 17 hours at 65°C in an Agilent hybridization oven (G2545A, Agilent Technologies) set to 10 rpm in a final concentration of 1x GEx Hybridization Buffer HI-RPM (Agilent Technologies). Arrays were washed and dried out using a centrifuge according to manufacturer's instructions (One-Color Microarray-Based Gene Expression Analysis, Agilent Technologies). Arrays were scanned at 5 mm resolution on an Agilent DNA Microarray Scanner (G2565BA, Agilent Technologies) using the default settings for 4x44k format one-color arrays. Images provided by the scanner were analyzed using Feature Extraction software v10.1.1.1 (Agilent Technologies).
Based on statistical outliers criteria implemented in the Bioconductor package "ArrayQualityMetrics" there were no statistical outliers in terms of MA plots, dendrogram or boxplots.
For the normalization, Agilent Processed Signal (Agilent Feature Extraction Software) was standardized across arrays using quantile normalization .
Differential gene expression was carried out using the limma  package from Bioconductor. This approach fits a linear model to all data considering as fixed effect the type of mice and using the array as blocking variable. Limma calculates moderated t-statistics, adding to the error term some information on the variance of all genes, solving the typical microarray problem of small sample size (2 in one of the experimental groups). Multiple testing adjustment of p-values was done according to Benjamini and Hochberg methodology .
The microarrays data of this study have been deposited in the Gene Expression Omnibus database under accession number GSE33647.
Ingenuity Pathway Analysis
Functional and Canonical Pathway analyses of specific gene datasets were generated through the use of Ingenuity Pathway Analysis (Ingenuity Systems®, http://www.ingenuity.com).
The Functional analysis identified functions and/or diseases that were most significant to the dataset. Genes from the dataset that were associated with biological functions and/or diseases in the Ingenuity knowledge base were considered for the analysis. B-H Multiple Testing Correction p-value test was used to calculate a p-value determining the probability that each biological function and/or disease assigned to the data set is due to chance alone.
Genes associated with a canonical pathway in the Ingenuity knowledge database were considered for the analysis. The significance of the association between the dataset and the canonical pathway was measured in two ways: 1) A ratio of the number of genes from the dataset that map to the pathway divided by the total number of molecules that exist in the canonical pathway is displayed. 2) B-H Multiple Testing Correction p-value test was used to calculate a p-value determining the probability that the association between the genes in the dataset and the canonical pathway is explained by chance alone. IPA identified significant networks, top functions and canonical pathways associated with the differentially expressed genes for each comparison analyzed.
FatiGO and FatiScan analysis
Enrichment analysis and gene set analysis was carried out for the Gene Ontology terms, InterPro terms and for the KEGG Pathways using FatiGO  and FatiScan  integrated in Babelomics suite .
FatiGO is a procedure to extract several functional terms as Gene Ontology (GO) terms, InterPro annotation or KEGG Pathways that are significantly over-or under-represented in sets of genes within the context of a genome-scale experiment. This resource is used to detect relevant functional terms for a group of genes with respect to a set of genes of reference (typically the rest of genes). The terms are considered to be relevant by the application of a Fisher's exact test that considers the multiple-testing nature of the statistical contrast performed.
FatiScan is a gene set analysis that detects significantly up-or down-regulated blocks of functionally related genes in lists of genes ordered by differential expression. FatiScan can search blocks of genes that are functionally related by different criteria such as gene ontology terms, Kyoto Encyclopedia of Genes and Genomes pathways, and others. The core of the method proposed is based on an algorithm to test whether a set of genes, labeled with terms (biological information), contain significant enrichments on one or several of these terms with respect to another set of genes of reference. FatiScan uses a Fisher's exact test for 2 × 2 contingency tables for comparing two groups of genes and extracting a list of GO terms whose distribution among the groups is significantly different. Given that many GO terms are simultaneously tested, the results of the test are corrected for multiple testing to obtain an adjusted p-value. FatiScan returns adjusted p-values based on False Discovery Rate (FDR) method [39, 40].
GO and InterPro annotation for the genes in the microarray where taken from Ensembl 56 release and KEGG Pathways from the KEGG web page.
RNA preparation and real-time quantitative PCR for validation of the microarray
Selected cDNA samples (n = 8-11) were quantified by real time quantitative PCR (qRT-PCR). Total RNA was isolated from islets samples according to the manufacturer's instructions following same protocols used for microarray RNA samples. Complimentary DNA was generated from 500 ng of RNA using M-MLV reverse transcriptase and master mix (Promega) in a 20 μl reaction with 2.5 mM MgCl2, 1.25 mM dNTPs and 5 μg/ml random hexamers at 37°C for 1 hour. cDNA was diluted 75 fold and 5 μl of diluted cDNA was used in a 12 μl real time PCR reaction using TaqMan primers and probes or SYBR green reagent (Applied Biosystems) according to manufacturers instructions. Reactions were run in duplicate for each sample and quantified in the ABI Prism 7900 sequence detection system (Applied Biosystems). Data was normalised to 18s rRNA. Primer sequences are shown in Additional file 1 Table S2.
Immunohistochemistry of TGF-beta
Fixed tissue sections (4 μm) were dehydrated by graded ethanol's and xylene, and then embedded in paraffin (2-3 sections per animal, 3-4 animal per group). The sections were deparaffinized and rehydrated. Sections were incubated with anti-TGF-β(SantaCruz Biotechnology, INC.). Sections were incubated with a biotinylated anti-IgG (Vector Laboratories) and incubated with the avidin-biotin-peroxidase complex (Vector Laboratories). 3, 3'-diaminobenzidine (DAB) substrate (Sigma-Aldrich®) was used as the chromogen. Some samples were incubated without primary antibody as negative controls. The stained sections were imaged with a light microscope Zeiss Standard 25.
We thank Janice Carter, Daniel Hart, Sergio Ferreiro and Davinia Hernandez for their work. We thank the funding bodies that have support research within our laboratories leading to this work: Diabetes UK, the MRC programme grant, MCINN BFU2009-10006 and BFU2008-04901-C03-03, CCG10-URJC/BIO-560 and L'Oreal-Unesco.
- Martin BC, Warram JH, Krolewski AS, Bergman RN, Soeldner JS, Kahn CR: Role of glucose and insulin resistance in development of type 2 diabetes mellitus: results of a 25-year follow-up study. Lancet. 1992, 340 (8825): 925-929. 10.1016/0140-6736(92)92814-V.View ArticlePubMedGoogle Scholar
- Prentki M, Nolan CJ: Islet beta cell failure in type 2 diabetes. J Clin Invest. 2006, 116 (7): 1802-1812. 10.1172/JCI29103.View ArticlePubMedPubMed CentralGoogle Scholar
- Rieck S, Kaestner KH: Expansion of beta-cell mass in response to pregnancy. Trends Endocrinol Metab. 2010, 21 (3): 151-158. 10.1016/j.tem.2009.11.001.View ArticlePubMedGoogle Scholar
- Doria A, Abumrad NA: Genome-wide associations and metabolic disease: the big revolution. Curr Opin Clin Nutr Metab Care. 2008, 11 (4): 363-365. 10.1097/MCO.0b013e32830460d7.View ArticlePubMedGoogle Scholar
- Saxena R, Voight BF, Lyssenko V, Burtt NP, de Bakker PI, Chen H, Roix JJ, Kathiresan S, Hirschhorn JN, Daly MJ, et al: Genome-wide association analysis identifies loci for type 2 diabetes and triglyceride levels. Science. 2007, 316 (5829): 1331-1336.View ArticlePubMedGoogle Scholar
- Scott LJ, Mohlke KL, Bonnycastle LL, Willer CJ, Li Y, Duren WL, Erdos MR, Stringham HM, Chines PS, Jackson AU, et al: A genome-wide association study of type 2 diabetes in Finns detects multiple susceptibility variants. Science. 2007, 316 (5829): 1341-1345. 10.1126/science.1142382.View ArticlePubMedPubMed CentralGoogle Scholar
- Sladek R, Rocheleau G, Rung J, Dina C, Shen L, Serre D, Boutin P, Vincent D, Belisle A, Hadjadj S, et al: A genome-wide association study identifies novel risk loci for type 2 diabetes. Nature. 2007, 445 (7130): 881-885. 10.1038/nature05616.View ArticlePubMedGoogle Scholar
- Steinthorsdottir V, Thorleifsson G, Reynisdottir I, Benediktsson R, Jonsdottir T, Walters GB, Styrkarsdottir U, Gretarsdottir S, Emilsson V, Ghosh S, et al: A variant in CDKAL1 influences insulin response and risk of type 2 diabetes. Nat Genet. 2007, 39 (6): 770-775. 10.1038/ng2043.View ArticlePubMedGoogle Scholar
- Zeggini E, Weedon MN, Lindgren CM, Frayling TM, Elliott KS, Lango H, Timpson NJ, Perry JR, Rayner NW, Freathy RM, et al: Replication of genome-wide association signals in UK samples reveals risk loci for type 2 diabetes. Science. 2007, 316 (5829): 1336-1341. 10.1126/science.1142364.View ArticlePubMedPubMed CentralGoogle Scholar
- Spiegelman BM: PPAR-gamma: adipogenic regulator and thiazolidinedione receptor. Diabetes. 1998, 47 (4): 507-514. 10.2337/diabetes.47.4.507.View ArticlePubMedGoogle Scholar
- Rosen ED, Kulkarni RN, Sarraf P, Ozcan U, Okada T, Hsu CH, Eisenman D, Magnuson MA, Gonzalez FJ, Kahn CR, et al: Targeted elimination of peroxisome proliferator-activated receptor gamma in beta cells leads to abnormalities in islet mass without compromising glucose homeostasis. Mol Cell Biol. 2003, 23 (20): 7222-7229. 10.1128/MCB.23.20.7222-7229.2003.View ArticlePubMedPubMed CentralGoogle Scholar
- Bruning JC, Winnay J, Bonner-Weir S, Taylor SI, Accili D, Kahn CR: Development of a novel polygenic model of NIDDM in mice heterozygous for IR and IRS-1 null alleles. Cell. 1997, 88 (4): 561-572. 10.1016/S0092-8674(00)81896-6.View ArticlePubMedGoogle Scholar
- Hull RL, Kodama K, Utzschneider KM, Carr DB, Prigeon RL, Kahn SE: Dietary-fat-induced obesity in mice results in beta cell hyperplasia but not increased insulin release: evidence for specificity of impaired beta cell adaptation. Diabetologia. 2005, 48 (7): 1350-1358. 10.1007/s00125-005-1772-9.View ArticlePubMedGoogle Scholar
- Parsons JA, Brelje TC, Sorenson RL: Adaptation of islets of Langerhans to pregnancy: increased islet cell proliferation and insulin secretion correlates with the onset of placental lactogen secretion. Endocrinology. 1992, 130 (3): 1459-1466. 10.1210/en.130.3.1459.PubMedGoogle Scholar
- Sachdeva MM, Stoffers DA: Minireview: Meeting the demand for insulin: molecular mechanisms of adaptive postnatal beta-cell mass expansion. Mol Endocrinol. 2009, 23 (6): 747-758. 10.1210/me.2008-0400.View ArticlePubMedPubMed CentralGoogle Scholar
- Medina-Gomez G, Gray SL, Yetukuri L, Shimomura K, Virtue S, Campbell M, Curtis RK, Jimenez-Linan M, Blount M, Yeo GSH, et al: PPAR gamma 2 prevents lipotoxicity by controlling adipose tissue expandability and peripheral lipid metabolism. Plos Genetics. 2007, 3 (4): e64-10.1371/journal.pgen.0030064.View ArticlePubMedPubMed CentralGoogle Scholar
- Medina-Gomez G, Yetukuri L, Velagapudi V, Campbell M, Blount M, Jimenez-Linan M, Ros M, Oresic M, Vidal-Puig A: Adaptation and failure of pancreatic beta cells in murine models with different degrees of metabolic syndrome. Disease Models & Mechanisms. 2009, 2: (11-12):582-592. 10.1242/dmm.002295.View ArticleGoogle Scholar
- Medina-Gomez G, Gray SL, Yetukuri L, Shimomura K, Virtue S, Campbell M, Curtis RK, Jimenez-Linan M, Blount M, Yeo GS, et al: PPAR gamma 2 prevents lipotoxicity by controlling adipose tissue expandability and peripheral lipid metabolism. PLoS Genet. 2007, 3 (4): e64-10.1371/journal.pgen.0030064.View ArticlePubMedPubMed CentralGoogle Scholar
- Cozar-Castellano I, Fiaschi-Taesch N, Bigatel TA, Takane KK, Garcia-Ocana A, Vasavada R, Stewart AF: Molecular control of cell cycle progression in the pancreatic beta-cell. Endocr Rev. 2006, 27 (4): 356-370. 10.1210/er.2006-0004.View ArticlePubMedGoogle Scholar
- Weyrich P, Staiger H, Stancakova A, Schafer SA, Kirchhoff K, Ullrich S, Ranta F, Gallwitz B, Stefan N, Machicao F, et al: Common polymorphisms within the NR4A3 locus, encoding the orphan nuclear receptor Nor-1, are associated with enhanced beta-cell function in non-diabetic subjects. BMC Med Genet. 2009, 10: 77.View ArticlePubMedPubMed CentralGoogle Scholar
- Gradwohl G, Dierich A, LeMeur M, Guillemot F: neurogenin3 is required for the development of the four endocrine cell lineages of the pancreas. Proc Natl Acad Sci USA. 2000, 97 (4): 1607-1611. 10.1073/pnas.97.4.1607.View ArticlePubMedPubMed CentralGoogle Scholar
- Willer CJ, Speliotes EK, Loos RJ, Li S, Lindgren CM, Heid IM, Berndt SI, Elliott AL, Jackson AU, Lamina C, et al: Six new loci associated with body mass index highlight a neuronal influence on body weight regulation. Nat Genet. 2009, 41 (1): 25-34. 10.1038/ng.287.View ArticlePubMedGoogle Scholar
- Sarruf DA, Iankova I, Abella A, Assou S, Miard S, Fajas L: Cyclin D3 promotes adipogenesis through activation of peroxisome proliferatoractivated receptor gamma. Mol Cell Biol. 2005, 25 (22): 9985-9995. 10.1128/MCB.25.22.9985-9995.2005.View ArticlePubMedPubMed CentralGoogle Scholar
- Miana M, de Las Heras N, Rodriguez C, Sanz-Rosa D, Martin-Fernandez B, Mezzano S, Lahera V, Martinez-Gonzalez J, Cachofeiro V: Effect of eplerenone on hypertension-associated renal damage in rats: potential role of peroxisome proliferator activated receptor gamma (PPARgamma). J Physiol Pharmacol. 2011, 62 (1): 87-94.PubMedGoogle Scholar
- Wang W, Liu F, Chen N: Peroxisome proliferator-activated receptorgamma (PPAR-gamma) agonists attenuate the profibrotic response induced by TGF-beta1 in renal interstitial fibroblasts. Mediators Inflamm. 2007, 2007: 62641.PubMedPubMed CentralGoogle Scholar
- Thomas HE, Trapani JA, Kay TW: The role of perforin and granzymes in diabetes. Cell Death Differ. 2010, 17 (4): 577-585. 10.1038/cdd.2009.165.View ArticlePubMedGoogle Scholar
- Bonfleur ML, Vanzela EC, Ribeiro RA, de Gabriel Dorighello G, de Franca Carvalho CP, Collares-Buzato CB, Carneiro EM, Boschero AC, de Oliveira HC: Primary hypercholesterolaemia impairs glucose homeostasis and insulin secretion in low-density lipoprotein receptor knockout mice independently of high-fat diet and obesity. Biochim Biophys Acta. 2010, 1801 (2): 183-190. 10.1016/j.bbalip.2009.10.012.View ArticlePubMedGoogle Scholar
- Joyce CW, Wagner EM, Basso F, Amar MJ, Freeman LA, Shamburek RD, Knapper CL, Syed J, Wu J, Vaisman BL, et al: ABCA1 overexpression in the liver of LDLr-KO mice leads to accumulation of pro-atherogenic lipoproteins and enhanced atherosclerosis. J Biol Chem. 2006, 281 (44): 33053-33065. 10.1074/jbc.M604526200.View ArticlePubMedGoogle Scholar
- Wiseman DA, Kalwat MA, Thurmond DC: Stimulus-induced Snitrosylation of Syntaxin 4 impacts insulin granule exocytosis. J Biol Chem. 2011, 286 (18): 16344-16354. 10.1074/jbc.M110.214031.View ArticlePubMedPubMed CentralGoogle Scholar
- Pan X, Dai Y, Li X, Niu N, Li W, Liu F, Zhao Y, Yu Z: Inhibition of arsenic induced-rat liver injury by grape seed exact through suppression of NADPH oxidase and TGF-beta/Smad activation. Toxicol Appl Pharmacol. 2011, 254 (3): 323-31. 10.1016/j.taap.2011.04.022.View ArticlePubMedGoogle Scholar
- Wijesekara N, Dai FF, Hardy AB, Giglou PR, Bhattacharjee A, Koshkin V, Chimienti F, Gaisano HY, Rutter GA, Wheeler MB: Beta cell-specific Znt8 deletion in mice causes marked defects in insulin processing, crystallisation and secretion. Diabetologia. 2010, 53 (8): 1656-1668. 10.1007/s00125-010-1733-9.View ArticlePubMedGoogle Scholar
- Xu K, Zha M, Wu X, Yu Z, Yu R, Xu X, Chen H, Yang T: Association between rs13266634 C/T polymorphisms of solute carrier family 30 member 8 (SLC30A8) and type 2 diabetes, impaired glucose tolerance, type 1 diabetes--a meta-analysis. Diabetes Res Clin Pract. 2011, 91 (2): 195-202. 10.1016/j.diabres.2010.11.012.View ArticlePubMedGoogle Scholar
- Bolstad BM, Irizarry RA, Astrand M, Speed TP: A comparison of normalization methods for high density oligonucleotide array data based on variance and bias. Bioinformatics. 2003, 19 (2): 185-193. 10.1093/bioinformatics/19.2.185.View ArticlePubMedGoogle Scholar
- Wettenhall JM, Smyth GK: limmaGUI: a graphical user interface for linear modeling of microarray data. Bioinformatics. 2004, 20 (18): 3705-3706. 10.1093/bioinformatics/bth449.View ArticlePubMedGoogle Scholar
- Hochberg Y, Benjamini Y: More powerful procedures for multiple significance testing. Stat Med. 1990, 9 (7): 811-818. 10.1002/sim.4780090710.View ArticlePubMedGoogle Scholar
- Al-Shahrour F, Diaz-Uriarte R, Dopazo J: FatiGO: a web tool for finding significant associations of Gene Ontology terms with groups of genes. Bioinformatics. 2004, 20 (4): 578-580. 10.1093/bioinformatics/btg455.View ArticlePubMedGoogle Scholar
- Al-Shahrour F, Minguez P, Tarraga J, Medina I, Alloza E, Montaner D, Dopazo J: FatiGO +: a functional profiling tool for genomic data. Integration of functional annotation, regulatory motifs and interaction data with microarray experiments. Nucleic Acids Res. 2007, W91-96. 35 Web ServerGoogle Scholar
- Medina I, Carbonell J, Pulido L, Madeira SC, Goetz S, Conesa A, Tarraga J, Pascual-Montano A, Nogales-Cadenas R, Santoyo J, et al: Babelomics: an integrative platform for the analysis of transcriptomics, proteomics and genomic data with advanced functional profiling. Nucleic Acids Res. 2010, W210-213. 38 Web ServerGoogle Scholar
- Benjamini Y, Hochberg Y: Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society. 1995, 57 (1): 289-300.Google Scholar
- Benjamini Y, Drai D, Elmer G, Kafkafi N, Golani I: Controlling the false discovery rate in behavior genetics research. Behav Brain Res. 2001, 125 (1-2): 279-284. 10.1016/S0166-4328(01)00297-2.View ArticlePubMedGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1755-8794/4/86/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.