Assessment of glucose transporter gene expression using the polymerase chain reaction.

We used a novel adaptation of the polymerase chain reaction to examine relative levels of mRNA encoding two members of the facilitative glucose transporter gene family, the GLUT1 or erythrocyte/HepG2/brain isoform and the GLUT4 or insulin-regulatable isoform. The method was fast (vs. hybridization methods), required no specific probe, and used total RNA samples of less than 1 microgram. Taking advantage of regions of structural similarity and differences between the two isoforms, we designed a single set of oligonucleotide primers capable of amplifying both GLUT1 and GLUT4 cDNAs such that their respective products could be resolved on the basis of a 12 base pair size differential. Hence, reverse transcription and complementary DNA amplification could be carried out for both transcripts using identical primers in the same reaction tube. Using this methodology, we examined the relative amounts of GLUT4 and GLUT1 mRNAs in several rat tissues. As expected based on prior reports using Northern analysis, rat brain contained only GLUT1 mRNA and skeletal muscle contained a large predominance of GLUT4 mRNA. Both isoform mRNAs were found in adipose tissue whereas adipose cells, heart and diaphragm contained predominantly GLUT4 mRNA. Induction of diabetes with streptozocin decreased the GLUT4 to GLUT1 ratio in adipose tissue 4-fold and 24 h of insulin treatment of the diabetic rats increased this ratio 9- to 10-fold. Insulin treatment of normal rats increased this ratio by 70%. Hindlimb skeletal muscle GLUT4 mRNA was quantified in diabetic and insulin-treated diabetic rats as a function of brain GLUT1 mRNA added as an internal standard. Using this methodology, no significant difference in muscle GLUT4 mRNA was noted as a result of 24 h of insulin therapy. In summary, quantitative PCR may be used to compare mRNA levels encoding specific members of a gene family either within given cells or tissues or as affected by physiological perturbations. Subject to certain limitations discussed within, this methodology may be useful in future measurements of glucose transporter mRNA, especially when only small tissue or cell samples are available.