Gene-expression measurement: variance-modeling considerations for robust data analysis.

System-wide measurements of gene expression by DNA microarray and, more recently, RNA-sequencing strategies have become de facto tools of modern biology and have led to deep understanding of biological mechanisms and pathways. However, analyses of the measurements have often ignored statistically robust methods that account for variance, resulting in misleading biological interpretations.

Multi-tissue, selective PPARγ modulation of insulin sensitivity and metabolic pathways in obese rats.

Peroxisome proliferator-activated receptor-γ (PPARγ) ligands, including the insulin-sensitizing thiazolidinedione drugs, transcriptionally regulate hundreds of genes. Little is known about the relationship between PPARγ ligand-specific modulation of cellular mechanisms and insulin sensitization. We characterized the insulin sensitivity and multitissue gene expression profiles of lean and insulin-resistant, obese Zucker rats untreated or treated with one of four PPARγ ligands (pioglitazone, rosiglitazone, troglitazone, and AG-035029). We analyzed the transcriptional profiles of adipose tissue, skeletal muscle, and liver from the rats and determined whether ligand treatment insulin-sensitizing potency was related to ligand treatment-induced alteration of functional pathways. Ligand treatments improved insulin sensitivity in obese rats to varying degrees. Adipose tissue profiles revealed ligand treatment-selective modulation of inflammatory and branched-chain amino acid (BCAA) metabolic pathways, which correlated with ligand treatment-specific insulin-sensitizing potency. Skeletal muscle profiles showed that obese rats exhibited elevated expression of adipocyte and slow-twitch fiber markers, which further increased after ligand treatment, but the magnitude of the treatment-induced changes was not correlated with insulin sensitization. Although PPARγ ligand treatments heterogeneously improved dysregulated expression of cholesterol and fatty acid biosynthetic pathways in obese rat liver, these alterations were not correlated with ligand insulin-sensitizing potency. PPARγ ligand treatment-specific insulin-sensitizing potency correlated with modulation of adipose tissue inflammatory and BCAA metabolic pathways, suggesting a functional relationship between these pathways and whole body insulin sensitivity. Other PPARγ ligand treatment-induced functional pathway changes were detected in adipose tissue, skeletal muscle, and liver profiles but were not related to degree of insulin sensitization.

PPARgamma activation in adipocytes is sufficient for systemic insulin sensitization.

Although peroxisome proliferator-activated receptor gamma (PPARgamma) agonists such as thiazolidinediones (TZDs) are widely used to treat type 2 diabetes, how its activation in individual tissues contributes to TZD's therapeutic action remains controversial. As TZDs are known to have receptor-independent effects, we sought to establish gain-of-function animal models to delineate the receptor's insulin-sensitizing actions. Unexpectedly, we find that selective activation of PPARgamma in adipocytes, but not in macrophages, is sufficient for whole-body insulin sensitization equivalent to systemic TZD treatment. In addition to improved adipokine, inflammatory, and lipid profiles, PPARgamma activation in mature adipocytes normalizes serum insulin without increased adipogenesis. Co-culture studies indicated that PPARgamma-activated adipocytes broadly suppress induction of inflammatory cytokines and C-X-C family chemokines in macrophages. Collectively, these data describe an "adipocentric" model in which adipose activation of PPARgamma is sufficient for complete insulin sensitization and suggest a specific application for fat selective PPARgamma modulators in diabetic therapy.

Chemoattractant signaling between tumor cells and macrophages regulates cancer cell migration, metastasis and neovascularization.

Tumor-associated macrophages are known to influence cancer progression by modulation of immune function, angiogenesis, and cell metastasis, however, little is known about the chemokine signaling networks that regulate this process. Utilizing CT26 colon cancer cells and RAW 264.7 macrophages as a model cellular system, we demonstrate that treatment of CT26 cells with RAW 264.7 conditioned medium induces cell migration, invasion and metastasis. Inflammatory gene microarray analysis indicated CT26-stimulated RAW 264.7 macrophages upregulate SDF-1alpha and VEGF, and that these cytokines contribute to CT26 migration in vitro. RAW 264.7 macrophages also showed a robust chemotactic response towards CT26-derived chemokines. In particular, microarray analysis and functional testing revealed CSF-1 as the major chemoattractant for RAW 264.7 macrophages. Interestingly, in the chick CAM model of cancer progression, RAW 264.7 macrophages localized specifically to the tumor periphery where they were found to increase CT26 tumor growth, microvascular density, vascular disruption, and lung metastasis, suggesting these cells home to actively invading areas of the tumor, but not the hypoxic core of the tumor mass. In support of these findings, hypoxic conditions down regulated CSF-1 production in several tumor cell lines and decreased RAW 264.7 macrophage migration in vitro. Together our findings suggest a model where normoxic tumor cells release CSF-1 to recruit macrophages to the tumor periphery where they secrete motility and angiogenic factors that facilitate tumor cell invasion and metastasis.

Diet-induced hepatocellular carcinoma in genetically predisposed mice.

Hepatocellular carcinoma (HCC) is one of the leading causes of cancer death worldwide, with approximately 70% of cases resulting from hepatitis B and C viral infections, aflatoxin exposure, chronic alcohol use or genetic liver diseases. The remaining approximately 30% of cases are associated with obesity, type 2 diabetes and related metabolic diseases, although a direct link between these pathologies and HCCs has not been established. We tested the long-term effects of high-fat and low-fat diets on males of two inbred strains of mice and discovered that C57BL/6J but not A/J males were susceptible to non-alcoholic steatohepatitis (NASH) and HCC on a high-fat but not low-fat diet. This strain-diet interaction represents an important model for genetically controlled, diet-induced HCC. Susceptible mice showed morphological characteristics of NASH (steatosis, hepatitis, fibrosis and cirrhosis), dysplasia and HCC. mRNA profiles of HCCs versus tumor-free liver showed involvement of two signaling networks, one centered on Myc and the other on NFkappaB, similar to signaling described for the two major classes of HCC in humans. miRNA profiles revealed dramatically increased expression of a cluster of miRNAs on the X chromosome without amplification of the chromosomal segment. A switch from high-fat to low-fat diet reversed these outcomes, with switched C57BL/6J males being lean rather than obese and without evidence for NASH or HCCs at the end of the study. A similar diet modification may have important implications for prevention of HCCs in humans.