During the American Diabetes Association national conference this year in Philadelphia, scientists gathered to discuss recent advances in this field at a session on positive and negative elements in signal transduction. When all was said and done, some puzzle pieces were added, but questions remain.
Because diabetes essentially results from the body's failure to respond to insulin, some researchers are trying to understand the signal transduction pathways that translate the binding of insulin to its receptor into cellular activity changes. The insulin receptor's cytoplasmic portion contains an inherent tyrosine kinase activity that becomes active on extracellular insulin binding, leading to the receptor's autophosphorylation and transphosphorylation of the insulin receptor substrates (IRS-1-3).1 These proteins associate with the regulatory subunit of phosphoinositol-3-kinase (PI3K), activating the enzyme's catalytic subunit, which adds a phosphate group to the 3'-OH position of the inositol ring in inositol phospholipids.
The reaction's products activate 3'-phosphoinositide-dependent kinase (PDK1), which in turn activates protein kinase B (PKB)/Akt, a serine kinase. Activated Akt plays a central role in regulating the number of insulin responses such as glycogen synthesis, which leads to movement of the glucose transporter, GLUT4, from intracellular stores to the plasma membrane. The net effect of these two processes is that the cell, responding to insulin, begins importing and storing glucose. The critical role of Akt in glucose homeostasis and insulin responsiveness is evident from the fact that Akt2 knockout mice develop diabetes.2
|Courtesy of Peter Miller|
Huffman purified and sequenced this new protein by mass spectrometry but could not identify it, except to say that it corresponded to the predicted protein product of a gene that had been sequenced in the human genome project. Then, in January, Karen Reue, a mouse geneticist at the University of California, Los Angeles, described the positional cloning of lipin, a gene (fld) that is mutated in mice afflicted with fatty liver dystrophy (FLD).3 These fld mutant mice fail to develop normal adipose tissue, are insulin resistant, and are hyperlipidemic--that is, they have high circulating levels of triglycerides that return to normal levels at about the time of weaning. Huffman found that his unknown protein was identical to the product of this newly identified gene.
Subsequent work has shown that, though lipin was identified in immunoprecipitated material obtained with an Akt2 antibody, lipin and Akt2 are not members of a single macromolecular complex--indeed, it appears that the coimmunoprecipitation of these two proteins was serendipitous.
"We're really nervous here, because we'd hate to miss something really important," says Lawrence in an interview, "But we think it was a function of either cross-reactivity of the original antibodies with lipin, or a contaminant in that antibody preparation. ... That seems to make the most sense with the data we have right now."
Nevertheless, it is beyond doubt that lipin is specifically phosphorylated in response to insulin stimulation, by way of the mammalian target-of-rapamycin (mTOR) pathway, downstream of Akt, says Lawrence. This work shows that lipin is a novel player in the insulin receptor pathway, one that promotes glucose intolerance when absent in mice. Furthermore, this work suggests a new link between insulin and lipid metabolism, one of its downstream effects, he says. There are, in fact, a number of interesting similarities between mTOR deficiency and the phenotype of the fld mutant mice. "We think this connection between mTOR and lipin could actually be telling us something pretty important," he adds.
The next step in this project is to identify the kinase responsible for phosphorylating lipin, as well as the residues within lipin that are phosphorylated in response to insulin, to see if mutation of those sites can affect lipin function, Lawrence notes. Of course, it will not be possible to define the effect of lipin mutations until the protein's normal biochemical function is defined.
Two Negatives = One Positive
Other researchers want to understand the mechanism of glucose uptake by insulin-treated cells. Avirup Bose, in Michael P. Czech's laboratory at the University of Massachusetts Medical Center in Worcester, reported research on glucose transporter migration to the plasma membrane in response to insulin. Bose and Czech studied GLUT4 translocation in fat cells treated with either insulin or endothelin-1, another hormone that induces GLUT4 translocation, and demonstrated that these hormones stimulate GLUT4 accumulation at the plasma membrane using different signaling circuitry. Whereas insulin treatment leads to GLUT4 translocation through the PI3K pathway, endothelin-1 uses a pathway involving the heterotrimeric GTPase (G) protein, G*11, and the actin regulatory small G protein, ADP-ribosylation factor (Arf)-6. Interestingly, these pathways must converge at some point, because latrunculin B, which induces F-actin disassembly, inhibits GLUT4 translocation caused by either hormone. Thus, glucose transporter migration to the cell surface is mediated by the actin cytoskeleton.
Morris J. Birnbaum, a Howard Hughes Medical Institute investigator at the University of Pennsylvania who helped organize, but did not address the symposium, also studies glucose transport. His lab observed results that are highly complementary to those described by Bose and Czech, and the two groups published back-to-back articles detailing their findings.4,5 Despite similarities, the two papers are not identical. Birnbaum's group, for example, did not directly present evidence implicating actin mobilization in GLUT4 trafficking. In talking with The Scientist, he also notes that the question of whether Arf6's effects on the cytoskeleton are directly responsible for mediating GLUT4 translocation in this system remains unresolved. Thus, according to Birnbaum, the next question is to define the link between actin reorganization and movement of transport proteins.
For instance, Birnbaum notes that in some cases, actin functions to prevent vesicle movement to the plasma membranes, while in others, vesicles appear to migrate along actin filaments as if they were railroad lines. Nevertheless, Birnbaum concedes that most researchers probably agree that actin reorganization is critical to insulin-regulated GLUT4 migration, but that a small G protein, other than Arf6, might play an important role. One potential candidate is the small G protein TC10, recently implicated in insulin-stimulated glucose uptake by Jeffrey Pessin and Alan Saltiel.6
Despite the notable array of research presented at the meeting, many unanswered questions remain, making it clear that, even though scientists are working to understand them, signal transduction pathways remain complicated processes. Scott A. Summers, assistant professor of biochemistry and molecular biology at Colorado State University, and one of the session's co-chairs, summed up the situation. The symposium showed "that people are really doing good work, and they're continuing to ask the right questions, but there are just a lot more questions that have to be asked."
Birnbaum takes the point further: These models, defined in vitro, must be validated at the organismal level. "Diabetes is fundamentally a disease of the intact organism, and the big advances in terms of understanding the relevance of the cell biology of the disease ... are going to come by applying the cell biological advances to in vivo models."
1.P. Bevan, "Insulin signalling," Journal of Cell Science, 114:1429-30, 2001.
2.H. Cho et al., "Insulin resistance and a diabetes mellitus-like syndrome in mice lacking the protein kinase Akt2 (PKBß)," Science, 292:1728-31, June 1, 2001.
3M. Peterfy et al., "Lipodystrophy in the fld mouse results from mutation of a new gene encoding a nuclear protein, lipin," Nature Genetics, 27:121-4, January 2001.
4.A. Bose et al., "G*11 signaling through ARF6 regulates F-actin mobilization and GLUT4 glucose transporter translocation to the plasma membrane," Molecular and Cellular Biology, 21:5262-75, August 2001.
5.J.T.R. Lawrence, M.J. Birnbaum, "ADP-ribosylation factor 6 delinea separate pathways used by endothelin 1 and insulin for stimulating glucose uptake in 3T3-L1 adipocytes," Molecular and Cellular Biology, 21:5276-85, August 2001.
6.S.H. Chiang et al., "Insulin-stimulated GLUT4 translocation requires the CAP-dependent activation of TC10," Nature, 410:944-8, April 19, 2001.