Lighting up insulin action.

Understanding the mechanism of insulin action remains one of the most important challenges in modern medical biology. Recent advances in cell imaging techniques, increased processing power of computers and the internet, and the introduction of novel fluorescent reagents such as green fluorescent proteins (GFPs) have revolutionized our ability to scrutinize insulin action by time-lapse microscopy at the single-cell level. This article outlines some of the advances made in the authors' laboratory, with particular reference to imaging the movements of the insulin-sensitive glucose transporter, GLUT4, and the generation of phosphoinositide lipids.

Role for the microtubule cytoskeleton in GLUT4 vesicle trafficking and in the regulation of insulin-stimulated glucose uptake.

Insulin stimulates glucose uptake into adipocytes by promoting the translocation of the glucose transporter isoform 4 (GLUT4) from intracellular vesicles to the plasma membrane. In 3T3-L1 adipocytes GLUT4 resides both in an endosomal pool, together with transferrin receptors, and in a unique pool termed 'GLUT4 storage vesicles' (GSVs), which excludes endosomal proteins. The trafficking of GLUT4 vesicles was studied in living 3T3-L1 adipocytes by time-lapse confocal microscopy of GLUT4 tagged with green fluorescent protein. GLUT4 vesicles exhibited two types of motion: rapid vibrations around a point and short (generally less than 10 microm) linear movements. The linear movements were completely blocked by incubation of the cells in the presence of microtubule-depolymerizing agents. This suggests that a subpopulation of GLUT4 vesicles can exhibit motor-driven movements along microtubules. Upon further examination, microtubule depolymerization inhibited insulin-stimulated glucose uptake and GLUT4 translocation to the plasma membrane by approx. 40%, but had no effect on insulin-induced translocation of the transferrin receptor to the plasma membrane from endosomes. We propose that an intact microtubule cytoskeleton may be required for optimal trafficking of GLUT4 present in the GSV pool, but not that resident in the endosomal pool.

Confocal imaging of the subcellular distribution of phosphatidylinositol 3,4,5-trisphosphate in insulin- and PDGF-stimulated 3T3-L1 adipocytes.

The activation of phosphatidylinositol 3-kinase (PI 3-kinase) and production of PtdIns(3,4,5)P(3) is crucial in the actions of numerous extracellular stimuli, including insulin-stimulated glucose uptake. Platelet-derived growth factor (PDGF) also stimulates PI 3-kinase, but only weakly promotes glucose uptake when compared with insulin. Insulin and PDGF have thus been proposed to have differential effects on the subcellular targeting of PI 3-kinase. However, owing to a lack of suitable methodologies, the subcellular localization of the PtdIns(3,4,5)P(3) generated has not been examined. The pleckstrin-homology (PH) domains of the nucleotide exchange factors, ADP-ribosylation factor nucleotide-binding-site opener (ARNO) and general receptor for 3-phosphoinositides (GRP1), which have a high affinity and specificity for PtdIns(3,4,5)P(3), were fused to green fluorescent protein and used to examine the subcellular localization of PtdIns(3,4,5)P(3) generation in living 3T3-L1 adipocytes. PtdIns(3,4,5)P(3) was produced almost exclusively in the plasma membrane in response to both agonists, although the response to insulin was greater in magnitude and occurred in considerably more cells. The results suggest that the greater ability of insulin to stimulate glucose uptake may be the result of its ability to generate significantly more plasma-membrane PtdIns(3, 4,5)P(3) than PDGF. ARNO and GRP1 are nucleotide exchange factors for the small GTP-binding protein ADP-ribosylation factor 6 (ARF6). The inability of a constitutively active GTPase-deficient mutant of ARF6 (ARF6-Q67L; Gln(67)-->Leu) to cause glucose transporter GLUT4 translocation suggests that activation of this pathway is not sufficient to cause GLUT4 translocation.

Protein kinase B stimulates the translocation of GLUT4 but not GLUT1 or transferrin receptors in 3T3-L1 adipocytes by a pathway involving SNAP-23, synaptobrevin-2, and/or cellubrevin.

An interaction of SNAP-23 and syntaxin 4 on the plasma membrane with vesicle-associated synaptobrevin-2 and/or cellubrevin, known as SNAP (soluble N-ethyl-maleimide-sensitive factor attachment protein) receptors or SNAREs, has been proposed to provide the targeting and/or fusion apparatus for insulin-stimulated translocation of the GLUT4 isoform of glucose transporter to the plasma membrane. By microinjecting 3T3-L1 adipocytes with the Clostridium botulinum toxin B or E, which proteolyzed synaptobrevin-2/cellubrevin and SNAP-23, respectively, we investigated the role of these SNAREs in GLUT4, GLUT1, and transferrin receptor trafficking. As expected, insulin stimulated the translocation of GLUT4, GLUT1, and transferrin receptors to the plasma membrane. By contrast, a constitutively active protein kinase B (PKB-DD) only stimulated a translocation of GLUT4 and not GLUT1 or the transferrin receptor. The GLUT4 response to PKB-DD was abolished by toxins B or E, whereas the insulin-evoked translocation of GLUT4 was inhibited by approximately 65%. These toxins had no significant effect on insulin-stimulated transferrin receptor appearance at the cell surface. Thus, insulin appears to induce GLUT4 translocation via two distinct routes, only one of which involves SNAP-23 and synaptobrevin-2/cellubrevin, and can be mobilized by PKB-DD. The PKB-, SNAP-23-, and synaptobrevin-2/cellubrevin-independent GLUT4 translocation pathway may involve movement through recycling endosomes, together with GLUT1 and transferrin receptors.

Identification of centaurin-alpha1 as a potential in vivo phosphatidylinositol 3,4,5-trisphosphate-binding protein that is functionally homologous to the yeast ADP-ribosylation factor (ARF) GTPase-activating protein, Gcs1.

Centaurin-alpha is a 46 kDa in vitro binding protein for the lipid second messenger PtdIns(3,4,5)P3. In this report we have addressed whether centaurin-alpha1, a human homologue of centaurin-alpha, binds PtdIns(3,4,5)P3 in vivo and furthermore, identified a potential physiological function for centaurin-alpha1. Using confocal microscopy of live PC12 cells, transiently transfected with a chimera of green fluorescent protein (GFP) fused to the N-terminus of centaurin-alpha1 (GFP-centaurin-alpha1), we demonstrated the rapid plasma membrane recruitment of cytosolic GFP-centaurin-alpha1 following stimulation with either nerve growth factor or epidermal growth factor. This recruitment was dependent on the centaurin-alpha1 pleckstrin homology domains and was blocked by the PtdIns(4,5)P2 3-kinase (PI 3-kinase) inhibitors wortmannin (100 nM) and LY294002 (50 microM), and also by co-expression with a dominant negative p85. Functionally, we demonstrated that centaurin-alpha1 could complement a yeast strain deficient in the ADP-ribosylation factor (ARF) GTPase-activating protein Gcs1; a complementation that was blocked by mutagenesis of conserved cysteine residues within the ARF GTPase-activating protein analogous domain of centaurin-alpha1. Taken together, our data demonstrated that centaurin-alpha1 could potentially function as an ARF GTPase-activating protein that, on agonist stimulation, was recruited to the plasma membrane possibly through an ability to interact with PtdIns(3,4,5)P3.

Nerve growth factor- and epidermal growth factor-stimulated translocation of the ADP-ribosylation factor-exchange factor GRP1 to the plasma membrane of PC12 cells requires activation of phosphatidylinositol 3-kinase and the GRP1 pleckstrin homology domain.

ADP-ribosylation factors (ARFs) are small GTP-binding proteins that are regulators of vesicle trafficking in eukaryotic cells. GRP1 is a member of a family of ARF guanine-nucleotide-exchange factors that binds in vitro the lipid second messenger phosphatidylinositol 3,4, 5-trisphosphate [PtdIns(3,4,5)P3]. In order to study the effects of PtdIns(3,4,5)P3 on the function of GRP1, we have cloned the human homologue of GRP1, encoding for a protein which is 98.8% identical to mouse brain GRP1. Human GRP1 binds, via its pleckstrin homology (PH) domain, the inositol head group of PtdIns(3,4,5)P3, inositol 1, 3,4,5-tetrakisphosphate [Ins(1,3,4,5)P4], with high affinity (Kd 32. 2+/-5.2 nM) and inositol phosphate specificity [Kd values for Ins(1, 3,4,5,6)P5, InsP6, Ins(1,3,4)P3 and Ins(1,4,5)P3: 283+/-32, >10000, >10000 and >10000 nM, respectively). Furthermore, GRP1 can accommodate addition of glycerol or diacetylglycerol to the 1-phosphate of Ins(1,3,4,5)P4, data that are consistent with its proposed role as a putative PtdIns(3,4,5)P3 receptor. To address whether GRP1 binds PtdIns(3,4,5)P3 in vivo, we have expressed a chimaera of green fluorescent protein (GFP) fused to the N-terminus of GRP1 in PC12 cells and, using confocal microscopy, examined its resultant localization in live cells. Stimulation with either nerve growth factor or epidermal growth factor (both at 100 ng/ml) results in a rapid, PH-domain dependent, translocation of GFP-GRP1 from the cytosol to the plasma membrane, which occurs with a time course that parallels the production of PtdIns(3,4,5)P3. This translocation is dependent on the activation of phosphatidylinositol 3-kinase, since it is inhibited by wortmannin (100 nM), LY294002 (50 microM) and by the co-expression with dominant negative p85. Taken together these data strongly suggest that GRP1 interacts in vivo with plasma membrane-located PtdIns(3,4,5)P3 and hence constitutes a true PtdIns(3,4,5)P3 receptor.

Insulin-dependent translocation of ARNO to the plasma membrane of adipocytes requires phosphatidylinositol 3-kinase.

ADP-ribosylation factors (ARFs) are small GTP-binding proteins that are regulators of vesicle trafficking in eukaryotic cells [1]. ARNO is a member of the family of guanine nucleotide exchange factors for ARFs which includes cytohesin-1 and GRP-1 [2] [3-5]. Members of this family contain a carboxy-terminal pleckstrin homology (PH) domain which, in the case of GRP-1, has been shown to bind the second messenger phosphatidylinositol 3,4,5-trisphosphate (PIP3) in preference to phosphatidylinositol 4,5-bisphosphate (PI(4,5)P2) and phosphatidylinositol 3,4-bisphosphate (PI(3,4)P2) in vitro [3,4]. Here, we show that recombinant ARNO has the binding characteristics of a PIP3 receptor and that this activity is restricted to the PH domain. When expressed in murine 3T3 L1 adipocytes, ARNO tagged using green fluorescent protein (GFP) is localised exclusively in the cytoplasm. Stimulation with insulin, however, causes a rapid (< 50 second) PH-domain-dependent translocation of GFP-ARNO to the plasma membrane. This translocation is blocked by the PI(4,5)P2 3-kinase (PI 3-kinase) inhibitors wortmannin and LY294002, and by co-expression with a dominant-negative p85 mutant, suggesting that the translocation is a consequence of insulin stimulation of PI 3-kinase. Our data strongly suggest that ARNO binds PIP3 in vivo and that this interaction causes a translocation of ARNO to the plasma membrane where it might activate ARF6 and regulate subsequent plasma membrane cycling events.

GLUT4 vesicle dynamics in living 3T3 L1 adipocytes visualized with green-fluorescent protein.

Insulin stimulates glucose uptake into its target cells by a process which involves the translocation of the GLUT4 isoform of glucose transporter from an intracellular vesicular compartment(s) to the plasma membrane. The step(s) at which insulin acts in the vesicle trafficking pathway (e.g. vesicle movement or fusion with the plasma membrane) is not known. We expressed a green-fluorescent protein-GLUT4 (GFP-GLUT4) chimaera in 3T3 L1 adipocytes. The chimaera was expressed in vesicles located throughout the cytoplasm and also close to the plasma membrane. Insulin promoted a substantial translocation of GFP-GLUT4 to the plasma membrane. Time-lapse confocal microscopy demonstrated that the majority of GFP-GLUT4-containing vesicles in the basal state were relatively static, as if tethered (or attached) to an intracellular structure. A proportion (approx. 5%) of the vesicles spontaneously lost their tether, and were observed to move rapidly within the cell. Other vesicles appear to be tethered only on one edge and were observed in a rapid stretching motion. The data support a model in which GLUT4-containing vesicles are tightly tethered to an intracellular structure(s), and indicate that a primary site of insulin action must be to release these vesicles, allowing them to then translocate to and fuse with the plasma membrane.

Inhibition of alanine:glyoxylate aminotransferase 1 dimerization is a prerequisite for its peroxisome-to-mitochondrion mistargeting in primary hyperoxaluria type 1.

Peroxisome-to-mitochondrion mistargeting of the homodimeric enzyme alanine:glyoxylate aminotransferase 1 (AGT) in the autosomal recessive disease primary hyperoxaluria type 1 (PH1) is associated with the combined presence of a normally occurring Pro(11)Leu polymorphism and a PH1-specific Gly170Arg mutation. The former leads to the formation of a novel NH2-terminal mitochondrial targeting sequence (MTS), which although sufficient to direct the import of in vitro-translated AGT into isolated mitochondria, requires the additional presence of the Gly170Arg mutation to function efficiently in whole cells. The role of this mutation in the mistargeting phenomenon has remained elusive. It does not interfere with the peroxisomal targeting or import of AGT. In the present study, we have investigated the role of the Gly170Arg mutation in AGT mistargeting. In addition, our studies have led us to examine the relationship between the oligomeric status of AGT and the peroxisomal and mitochondrial import processes. The results obtained show that in vitro-translated AGT rapidly forms dimers that do not readily exchange subunits. Although the presence of the Pro(11)Leu or Gly170Arg substitutions alone had no effect on dimerization, their combined presence abolished homodimerization in vitro. However, AGT containing both substitutions was still able to form heterodimers in vitro with either normal AGT or AGT containing either substitution alone. Expression of various combinations of normal and mutant, as well as epitope-tagged and untagged forms of AGT in whole cells showed that normal AGT rapidly dimerizes in the cytosol and is imported into peroxisomes as a dimer. This dimerization prevents mitochondrial import, even when the AGT possesses an MTS generated by the Pro(11)Leu substitution. The additional presence of the Gly170Arg substitution impairs dimerization sufficiently to allow mitochondrial import. Pharmacological inhibition of mitochondrial import allows AGT containing both substitutions to be imported into peroxisomes efficiently, showing that AGT dimerization is not a prerequisite for peroxisomal import.

Molecular basis of the variable mitochondrial and peroxisomal localisation of alanine-glyoxylate aminotransferase.

The molecular basis of the variable species-specific peroxisomal and/or mitochondrial targeting of the enzyme alanine-glyoxylate aminotransferase 1 (AGT) has been studied in human fibroblasts by confocal immunofluorescence microscopy after intranuclear microinjection of various human, rabbit, marmoset, and feline AGT cDNA constructs. The expression of full-length human and rabbit AGT cDNA led to an exclusively peroxisomal distribution of AGT. However, the distribution of feline and marmoset AGT depended on the cDNA construct injected. In both species, injection of the short cDNAs (from transcripts that occur naturally in marmoset liver but not in feline liver) led to an exclusively peroxisomal distribution. However, injection of the long cDNAs (from transcripts that occur naturally in both species) led to most of the AGT being targeted to the mitochondria and only a small, yet significant, fraction to the peroxisomes. Reintroduction of the 'ancestral' first potential translation initiation site into human AGT cDNA led to an 'ancestral' distribution of AGT (i.e. both mitochondrial and peroxisomal). Deletion of the second potential translation start site from the long feline cDNA led to a distribution that was almost entirely mitochondrial, which suggests that most peroxisomal AGT encoded by the long cDNA results from internal translation initiation from this site with the consequent loss of the N-terminal mitochondrial targeting sequence. Expression of rabbit cDNA and the short marmoset and feline cDNAs in cells selectively deficient in the import of peroxisomal matrix proteins showed that peroxisomal AGT in all these species is imported via the peroxisomal targeting sequence type 1 (PTS1) import pathway. The almost complete functional dominance of the N-terminal mitochondrial targeting sequence over the C-terminal PTS. which was not due to any direct interference of the former with peroxisomal import, was maintained even when the unusual PTS1 of AGT (KKL in human) was replaced by the prototypical PTS1 SKL. The results demonstrate that the major determinant of alanine-glyoxylate aminotransferase subcellular distribution in mammals is the presence or absence of the mitochondrial targeting sequence rather than the peroxisomal targeting sequence. Various strategies have arisen during the evolution of mammals to enable the exclusion of the mitochondrial targeting sequence from the newly synthesised polypeptide, all of which involve the use of alternative transcription and/or translation initiation sites.

Mammalian alanine/glyoxylate aminotransferase 1 is imported into peroxisomes via the PTS1 translocation pathway. Increased degeneracy and context specificity of the mammalian PTS1 motif and implications for the peroxisome-to-mitochondrion mistargeting of AGT in primary hyperoxaluria type 1.

Alanine/glyoxylate aminotransferase 1 (AGT) is peroxisomal in most normal humans, but in some patients with the hereditary disease primary hyperoxaluria type 1 (PH1), AGT is mislocalized to the mitochondria. In an attempt to identify the sequences in AGT that mediate its targeting to peroxisomes, and to determine the mechanism by which AGT is mistargeted in PH1, we have studied the intracellular compartmentalization of various normal and mutant AGT polypeptides in normal human fibroblasts and cell lines with selective deficiencies of peroxisomal protein import, using immunofluorescence microscopy after intranuclear microinjection of AGT expression plasmids. The results show that AGT is imported into peroxisomes via the peroxisomal targeting sequence type 1 (PTS1) translocation pathway. Although the COOH-terminal KKL of human AGT was shown to be necessary for its peroxisomal import, this tripeptide was unable to direct the peroxisomal import of the bona fide peroxisomal protein firefly luciferase or the reporter protein bacterial chloramphenicol acetyltransferase. An ill-defined region immediately upstream of the COOH-terminal KKL was also found to be necessary for the peroxisomal import of AGT, but again this region was found to be insufficient to direct the peroxisomal import of chloramphenicol acetyltransferase. Substitution of the COOH-terminal KKL of human AGT by the COOH-terminal tripeptides found in the AGTs of other mammalian species (SQL, NKL), the prototypical PTS1 (SKL), or the glycosomal PTS1 (SSL) also allowed peroxisomal targeting, showing that the allowable PTS1 motif in AGT is considerably more degenerate than, or at least very different from, that acceptable in luciferase. AGT possessing the two amino acid substitutions responsible for its mistargeting in PH1 (i.e., Pro11-->Leu and Gly170-->Arg) was targeted mainly to the mitochondria. However, AGTs possessing each amino acid substitution on its own were targeted normally to the peroxisomes. This suggests that Gly170-->Arg-mediated increased functional efficiency of the otherwise weak mitochondrial targeting sequence (generated by the Pro11-->Leu polymorphism) is not due to interference with the peroxisomal targeting or import of AGT.