Thiazolidinediones and their effects on glucose transporters.

Abstract

Recently, considerable attention has been paid to a promising new family of antidiabetic therapeutic agents, the thiazolidinediones. Several members of this family have been shown to have potent effects to lower hyperglycemia, hyperinsulinemia and hypertriglyceridemia in insulin resistant humans and animals. Even as one member of this family, troglitazone (Rezulin), has become available for clinical use in the US and Japan, we still do not fully understand the mechanisms by which these agents exert their effects. Shimaya and colleagues (1) have employed an insulin resistant model system, the obese Zucker rat, to investigate potential mechanisms of thiazolidinedione action. Treatment of obese animals with a new thiazolidinedione analog, YM268, greatly reduced fasting glucose and insulin levels and improved glucose tolerance. These workers confirmed the reduced glucose transporter 4 (GLUT4) protein expression in adipocytes of obese animals compared with lean littermates and found that YM268 treatment resulted in large increases in GLUT4 protein, along with a generalized increase in membrane protein. They suggest that this change can account, at least in part, for the demonstrated amelioration of insulin resistance after treatment. This, and related work, give rise to two questions. First, what is the ultimate significance of changes in GLUT4 expression in adipose tissue? Secondly, how might thiazolidinediones be acting to regulate glucose transporter expression? Effects of thiazolidinediones on adipose tissue glucose transport activity and transporter expression have been reported since the earliest papers on these agents. A common observation, regardless of the animal model or agent tested, is that glucose transport activity is increased, as is GLUT4 expression (2–6) following treatment. These changes occur together with reductions in fasting glucose and insulin levels, so it is uncertain if the effects on transport/transporters are direct or due to the change in the metabolic environment. Considering the demonstrated abilities of glucose and insulin to regulate glucose transport, it is difficult to separate the cause from the effect. Studies in cultured cell systems, where glucose and insulin levels can be tightly controlled, could help to resolve this issue. 3T3-L1 and 3T3-F442A adipocytes have been treated with a variety of thiazolidinediones, resulting in increases in transport activity and transporter expression (7–11). However, these treatments were performed during fibroblast differentiation into adipocytes and, since the appearance of GLUT4 is one marker of adipocyte differentiation, it is uncertain whether this result is due to augmented differentiation, or a more specific effect on transport which would have greater relevance to the therapeutic situation where adipocytes are already differentiated. Given the results showing that thiazolidinediones alone can drive adipocyte differentiation (7, 11), or augment that due to insulin (7), it is even more difficult to draw definitive conclusions from the majority of these studies. However, it has been shown that troglitazone treatment of 3T3-L1 cells following differentiation into adipocytes resulted in upregulation of basal glucose transport and GLUT1 expression (11). GLUT4 expression was unaltered, further suggesting that changes in GLUT4 may be indirect responses in adipose cells. In even the most obese individual (or animal), the portion of a glucose load utilized in adipose tissue is a modest fraction of the total. It is glucose uptake into muscle that represents the major portion of insulinstimulated glucose uptake and is the most significant site of insulin resistance. Thiazolidinedione treatment of insulin resistant rodents has also been shown to augment glucose uptake into muscle as well as increase GLUT4 expression (2–4, 6, 8, 12–14). Such changes could account, to a greater extent, for increases in whole body glucose disposal and would represent the most significant action of thiazolidinediones. Again, the question remains whether these are direct effects of thiazolidinediones, or due to changes in insulin and glucose levels. Fortunately, data in cultured muscle cells are helpful in resolving this issue. Troglitazone treatment of L6 myocytes (15) and isolated rat cardiac myocytes (16) increased glucose transport and the expression of both GLUT1 and GLUT4. This effect was independent of differentiation because it occurred even in L6 cells that were not differentiated into myotubes (15) or in BC3H-1 myocytes (with pioglitazone) (17) that are incapable of differentiation. It should be noted that the non-differentiated myocytes express only GLUT1 and it is this isoform that is upregulated by thiazolidinedione treatment. We have performed similar European Journal of Endocrinology (1997) 137 610–612 ISSN 0804-4643