Animal models used to identify differentially expressed liver genes
All experimental protocols were approved by the Ohio University Animal Care and Utilization Committee.
For the analysis of gene expression during the aging process, twelve male C57Bl/6J mice were fed standard chow (PMI Nutrition International Inc., Brentwood, MO, Prolab RMH3000) immediately following weaning. Mice were sacrificed at 35, 49, 56, 77, 133, 207, 403, 558 and 725 days of age. A portion of the left lateral lobe of the liver was placed in 4% paraformaldehyde and the remaining tissue was frozen in liquid nitrogen until processed for RNA isolation. For the analysis of normal versus diabetic liver gene expression, we utilized a mouse model of obesity-induced type 2 diabetes first described by Surwit  and replicated successfully in our laboratory and others [21, 22]. Mice were weaned at 3 weeks onto either standard chow or a high-fat diet (BioServe, Frenchtown, NJ. #F1850) for up to 26 weeks. Representative mice were sacrificed after 2, 4, 8, 16 and 26 weeks on the diet (35, 49, 77, 133 and 203 days of age) and liver tissue was isolated. Mice fed standard chow served as control animals. Type 2 diabetic mice were characterized as having > 200 mg/dl fasting blood glucose and at least a 2-fold increase in fasting plasma insulin compared to control mice.
For diet-reversal analysis, C57Bl/6J mice were weaned onto a standard chow (SC) diet and maintained for 47 weeks. A second set of mice were weaned onto a high-fat diet (HF) diet for 33 weeks. One half of the high-fat fed mice were maintained on the high-fat diet while the other half was switched to standard chow for a period of 14 weeks. Food and water were supplied ad libitum throughout the course of the studies. Following the 47 week feeding period, the animals were sacrificed and liver tissue isolated as described above.
Glucose and insulin levels
Blood-glucose levels were measured from blood taken from the tip of the tail of fasted (8 hr) mice using a One Touch glucometer (Life scan). All measurements occurred between 2:00 and 5:00 pm. Insulin concentrations were determined using the Ultrasensitive Rat Insulin ELISA kit (ALPCO, Windham, NH) as instructed by the manufacturer. Values were adjusted by a factor of 1.23 as determined by the manufacturer to correct for the species difference in cross-reactivity with the antibody.
Total RNA was isolated from whole liver using the RNA STAT-60 Total RNA/mRNA Isolation Reagent according to the manufacturer's instructions (Tel-Test, Friendswood, TX). Biotinylated cRNA hybridization target was prepared by a linear amplification method as described in the manufacturer's instructions for CodeLink Expression Bioarrays™ (Amersham Biosciences). The oligonucleotide probes were provided by the Codelink Uniset Mouse I Bioarray (Amersham, product code 300013). CodeLink Expression Bioarrays™ contain 10,000 oligonucleotide probes, each specific to a well-characterized mouse gene. Using the cRNA target, the hybridization reaction mixture containing the biotinylated target cRNA was loaded into array chambers for bioarray processing as described in the manufacturer's instructions for CodeLink Gene Expression BioarraysTM (Amersham Biosciences). Each sample was hybridized at 37°C to an individual microarray. Hybridization was detected with an avidinated fluorescent reagent, Streptavidin-Alexa Fluor ® 647 (Amersham). Processed arrays were scanned using a GenePix 4000B Microarray Scanner (Axon Instruments, Inc.). Array images were acquired using the Amersham CodeLink Analysis Software (Release 2.2). A significant difference in expression between samples was defined as a minimum of 2-fold change in expression values.
RNA was isolated from frozen liver as described above. cDNA transcripts were created using iScript cDNA Synthesis Kit (Cat# 170-8891, Bio-Rad Laboratories, Hercules, CA). Real Time RT-PCR was performed on the Bio-Rad iCycler iQ Real Time PCR Detection System (Cat# 170-8740, Bio-Rad Laboratories) using IQ SYBR Green Supermix (Cat# 170-8882). CIDE-A primers (Forward Primer: 5'CTCGGCTGTCTCAATGTCAA3'; Reverse Primer: 5'CCGCATAGACCAGGAACTGT3' were designed using Primer3 online software . Housekeeping genes, ACTG (actin gamma cytoplasmic) and GADPH (glyceraldehyde-III phosphate dehydrogenase) were utilized for normalization as described by Vandesompele et al. [40, 39]. Relative expression levels were then calculated using normalization factors derived from geNorm analysis of ACTG and GADPH using the delta-delta CT method .
Liver histology and image analysis
Liver tissues fixed in 4% paraformaldehyde were embedded in Tissue Path (Fisher Scientific, Pittsburgh, PA). Embedded tissue was sectioned for seven microns thickness, stained with H&E and evaluated by an automated light microscopy system, consisting of a 20 × lens with 0.5 NA, scanning stage, 3-color CCD camera and image analysis software and database (Icoria Inc, Research Triangle Park, NC). Images were taken at 0.64 micron resolution and were automatically assembled into montages. From 300 to 500 single images were captured to represent a single tissue section. These montages were then used for subsequent automated analysis of both gross anatomical features and fine tissue structures using automated pathology software [41, 42]. Tissue metrics included counts, area, density and size of hepatocyte nuclei, non-hepatocyte nuclei, percent intracellular and extracellular white space. The feature intracellular white space calculated the amount of white space within hepatic boundaries, the appearance of which is consistent with lipid accumulation in this study. Application of these methods resulted in distinct tissue profiles for all animals.
Total RNA (10 μg) from appropriate tissues was resolved by denaturing agarose gel electrophoresis, transferred to nylon membrane, hybridized with the [α-32P]dCTP-labeled mouse CIDE-A cDNA (Random Primed DNA Labeling Kit, Roche, Indianapolis, IN) and exposed to Bio-Max MR film (Eastman Kodak Co., Rochester, NY).
Liver and heart tissue (100 mg) was homogenized in 0.5 ml phosphate buffered saline containing 7.5 μl protease inhibitor cocktail (Sigma #P8340, St. Louis, MO). The samples were centrifuged for 5 min at 10,000 g. The supernatant was collected and protein concentration determined (Bio-Rad Laboratories #500-0006, Hercules, CA). Sixty μg of each extract was electrophoresed on a 12.5% SDS-polyacrylamide gel as described . Resolved proteins were transferred to a nitrocellulose membrane and immunoblotted using a rabbit anti-mouse CIDE-A polyclonal antibody (QED Bioscience Inc., San Diego, CA) as previously described .
Data are reported as dot plots or as mean (SD). Differences between two groups were assessed using the unpaired two-tailed Student's t test. For comparisons between multiple groups, ANOVA followed by Tukey's multiple-comparisons test was used. Analysis of correlations was done with Pearson correlation coefficients. Statistical significance was indicated by P value less than 0.05. SigmaStat statistics software (version 12.0; SPSS) was used for all calculations.