Lai-Har Chi
Back Home Up Next

Home
Russell D. Wolfinger
David Allison
Jenny Bryan
Lai-Har Chi
Jane Chang
Philip Dixon
Kent M. Eskridge
Deborah Glueck
David L. Gold
Susan G. Hilsenbeck
Lawrence Hunter
Rebecka Jornsten
Steen Knudsen
Laura Lazzeroni
Chen-Tuo Liao
Peter Munson
Dan Nettleton
Wei Pan
David M. Rocke
Grace S. Shieh
Lue Ping Zhao
Deepak Mav
Annette Molinaro

Identification of Gene Expression Changes in Skeletal Muscle from Diabetic and Normal Rats Treated by Oral Administration of Vanadyl Sulfate

Gail R. Willsky1, Lai-Har Chi1 Yulan Liang2 and Debbie C Crans3 
University at Buffalo (SUNY) Dept. of Biochemistry1 or Division of Statistics, Dept. of Social and Preventative Medicine2 and Colorado State University, Dept. of Chemistry3

Oral treatment with vanadium compounds alleviates the symptoms of diabetes in both rodents and humans. This study examined changes in global gene expression in muscle caused by STZ-induced diabetes and VOSO4 treatment of rats with STZ-induced diabetes. Expression data was first obtained from Normal (N), Diabetic (D), and Diabetic treated with VOSO4 (D/VOSO4) rats, with five animals in each group. Labeled cRNA from each animal was hybridized to a separate Affymetrix U34A GeneChip. Of the 8799 probe sets present on the chip, 5055 were expressed in one of our groups and included in the analysis. All possible group comparisons were made using a t test, and p values were obtained for each probe set using Affymetrix DMT 3.0. The decision threshold for a 20% false positive rate was 3.95 x 10-5. A p value less than this decision threshold value was observed for two genes. Gene expression in skeletal muscle from D rats was compared to that in N rats; in D/VOSO4 rats compared to N rats; and D/ VOSO4 rats compared to D rats. The 15 gene expression files from the three experimental groups were scaled using a target intensity of 2500 for normalization across samples. Pair-wise comparisons of each set were done using Affymetrix MAS 5.0. Twenty-five matrices were examined for each comparison. Probe sets with similar expression patterns in 80% of the matrices were selected. A total of 259 probe sets were identified as fitting the criteria. When gene expression in D/VOSO4 animals was compared to that in N animals only 6 probe sets were up regulated and 5 probe sets down regulated, demonstrating that VOSO4 treatment of D animals corrected many of the D induced changes in gene expression. There were 32 genes up regulated in D animals and down regulated in D/VOSO4 animals, with one gene showing the opposite expression pattern. These genes include metallothionein, 3-hydroxy-3-methylglutaryl-CoA synthase, and glutathione S transferase, which would be expected from the known interactions of vanadium compounds with cellular metabolism. Other genes identified could be divided into four major groups involving lipid metabolism, transport and trafficking, muscle structure, and signal transduction Hypoglycemia is a serious problem in the management of diabetes. Unlike treatment with insulin, treatment with VOSO4 does not cause symptoms of hypoglycemia. To understand the differential effect of VOSO4 treatment on N and D animals, the microarray data obtained was expanded to include a fourth group of five rats, Normal treated with VOSO4 (N/VOSO4). In comparing the gene expression of the N/VOSO4 and N animals only 35 probe sets were identified that were up regulated by VOSO4 treatment and one probe set was down regulated by the treatment. In addition, only one of those probe sets had been identified as being up regulated by diabetes and down regulated by VOSO4 treatment of diabetic animals. We performed statistical analysis utilizing two types of two way ANOVA models to estimate differential expression of genes between different animals (arrays). The first is designed for the estimation of the array (sample) treatment effects through the pair-wise comparisons of the four different groups. The second considered that the experimental treatment involved two factors, where being diabetic compared to normal was one factor and being treated with vanadium was the other. The results of both models consistently agree with each other and showed that there was a significant interaction term between treatment and disease as implied by our original data analysis. Both models also showed no significant differences when N animals were compared to D/VOSO4 animals, while other groups were significantly different from each other. These results demonstrate that VOSO4 treatment of diabetic animals compensates for specific changes in gene expression induced by D and restores global gene expression patterns towards that observed in N animals. In addition, comparison of gene expression in both N and D animals may help to elucidate the differential effect of VOSO4 treatment on these animals and help develop ways to avoid hypoglycemic episodes in diabetic patients.

 

Graybill Conference
June 18-20, 2003
University Park Holiday Inn
Fort Collins, CO 80526
www.stat.colostate.edu/graybillconference
email: hari@stat.colostate.edu Fax: (970)491-7895 Phone: (970)491-5269
Last Updated: Tuesday, May 13, 2003