High basal cell surface levels of fish GLUT4 are related to reduced sensitivity of insulin-induced translocation toward GGA and AS160 inhibition in adipocytes

Abstract

Glucose entry into cells is mediated by a family of facilitative transporter proteins (GLUTs). In mammals, GLUT4 is expressed in insulin-sensitive tissues and is responsible for the postprandial uptake of glucose. In fish, GLUT4 also mediates insulin-regulated glucose entry into cells but differs from mammalian GLUT4 in its affinity for glucose and in protein motifs known to be important for the traffic of GLUT4. In this study, we have characterized the intracellular and plasma membrane (PM) traffic of two orthologs of GLUT4 in fish, trout (btGLUT4) and salmon (okGLUT4), that do not share the amino terminal FQQI targeting motif of mammalian GLUT4. btGLUT4 (FQHL) and, to a lesser extent, okGLUT4 (FQQL) showed higher basal PM levels, faster traffic to the PM after biosynthesis, and earlier acquisition of insulin responsiveness than rat GLUT4. Furthermore, btGLUT4 showed a similar profile of internalization than rat GLUT4. Expression of the dominant-interfering AS160-4P mutant caused a significant decrease in the insulin-induced PM levels of okGLUT4 and rat GLUT4 and, to a lesser extent, of btGLUT4, suggesting that btGLUT4 has reduced retention into the IRC. Contrary to rat GLUT4 and okGLUT4, the presence of btGLUT4 at the PM under insulin-stimulated conditions was not affected by coexpression of a dominant-interfering GGA mutant. These data suggest that fish GLUT4 follow a different trafficking pathway to the PM compared with rat GLUT4 that seems to be relatively independent of GGA. These results indicate that the regulated trafficking characteristics of GLUT4 have been modified during evolution from fish to mammals.

Fig. 1.

The basal cell surface levels of fish GLUT4s are higher than those of rat GLUT4. A: differentiated 3T3-L1 adipocytes expressing either the rat GLUT4, okGLUT4 (salmon), or btGLUT4 (trout) were serum starved and incubated with or without insulin (100 nM, 30 min), as described in materials and methods. Data are presented as percentage (means ± SE) of cells showing a complete plasma membrane (PM) rim obtained by counting 100 cells/condition in 3 independent experiments. Statistical analysis was performed by unpaired t-test between each of the fish GLUT4s and the rat GLUT4 at basal conditions (*significance at P < 0.05). In addition, unpaired t-test for each GLUT4 construct was done between basal and insulin-stimulated conditions (#significance at P < 0.05). B: representative images of differentiated 3T3-L1 adipocytes expressing the btGLUT4-EGFP (green: a, d, g, and j) and the rat GLUT4-myc (red: b, e, h, and k) constructs. Immunofluorescence was performed as described in materials and methods; images were obtained by laser confocal microscopy with a Leica DMIRE2 microscope.

Fig. 2.

The elevated btGLUT4 basal cell surface levels are not due to reduced internalization. Differentiated 3T3-L1 adipocytes expressing either the rat GLUT4 or btGLUT4 (trout) were serum starved and incubated with insulin (100 nM, 30 min), labeled at 4°C with c-Myc antibody, and internalization followed at 37°C over time, as described in materials and methods. Data are presented as arbitrary units (means ± SE) of PM red fluorescence divided by total cell red fluorescence obtained from 12 cells/condition in 2 independent experiments. Statistical analysis was performed by paired t-test between btGLUT4 and the rat GLUT4 at each time point, but no significant differences were observed at P < 0.05.

Fig. 3.

btGLUT4 has impaired AS160-dependent sequestration in the insulin-responsive intracellular storage compartments (IRC). Differentiated 3T3-L1 adipocytes coexpressing either rat GLUT4 (ratG4), okGLUT4 (salmon; okG4), or btGLUT4 (trout; btG4) plus AS160-WT or AS160-4P were serum starved and incubated with or without insulin (100 nM, 30 min), as described in materials and methods. Data are presented as percentage (means ± SE) of cells showing a complete PM rim obtained by counting 100 AS160-positive cells/condition in 3 independent experiments. Statistical analysis was performed by unpaired t-test between each of the fish GLUT4s and the ratG4 cotransfected with the corresponding AS160 construct at basal conditions (*significance at P < 0.05). In addition, unpaired t-test for each GLUT4 construct cotransfected with AS160-4P was done against each corresponding AS160-WT group at basal or insulin-stimulated conditions, respectively (#significance at P < 0.05). Table (inset) shows the %inhibition of insulin stimulation of PM localization (fold change over basal) in AS160-4P- compared with AS160-WT-expressing cells, including the P values (those <0.05 are shown in boldface).

Fig. 4.

Fish GLUT4s acquire insulin-stimulated translocation earlier than rat GLUT4. Differentiated 3T3-L1 adipocytes expressing either ratG4, okG4, or btG4 were serum starved and incubated with or without insulin (100 nM, 30 min) to end the experiment at exactly 3, 6, 9, or 12 h after transfection, as described in materials and methods. Data are presented as percentage (means ± SE) of cells showing a complete PM rim obtained by counting 100 cells/condition in 3 independent experiments. Statistical analysis was performed by unpaired t-test between each of the fish GLUT4s and the ratG4 at basal and insulin-stimulated conditions (*significance at P < 0.05). In addition, unpaired t-test for each GLUT4 construct was performed between basal and insulin-stimulated conditions at each time point (#significance at P < 0.05).

Fig. 5.

btGLUT4 has reduced sensitivity to Golgi-localized γ-ear-containing Arf-binding protein (GGA) regulation. Differentiated 3T3-L1 adipocytes coexpressing either ratG4, okG4, or btG4 plus GGA-wild type (WT) or GGA-dominant interfering (DN) were serum starved and incubated with or without insulin (100 nM, 30 min) at 6 (A) or 24 h (B) after transfection, as described in materials and methods. Data are presented as percentage (means ± SE) of cells showing a complete PM rim obtained by counting 100 GGA-positive cells/condition in 3 independent experiments. A: *significant differences between each of the fish GLUT4s and ratG4 at insulin-stimulated conditions (P < 0.05). B: *significant differences between each of the fish GLUT4s and ratG4 at basal conditions (P < 0.05). In addition, a t-test for each GLUT4 construct cotransfected with GGA-DN was done against each corresponding GGA-WT group at basal or insulin-stimulated conditions, respectively (#significance at P < 0.05), but no significant differences were observed at 6 h after transfection. Tables (insets) show the %inhibition of insulin stimulation of PM localization (fold change over basal) in GGA-DN- compared with GGA-WT-expressing cells, including the P values (those <0.05 are shown in boldface). Representative images of differentiated 3T3-L1 adipocytes expressing the btGLUT4-EGFP (green: a, d, g, and j) and the DsRed-GGA-WT (red: b, e, h, and k) (C) constructs and the rat GLUT4-EGFP (green: a and d) and the DsRed-GGA-WT (red: b and e) (D) constructs. The images were obtained by laser confocal microscopy with a Leica DMIRE2 microscope.