- Email: [email protected]
Acylation-stimulating protein (ASP) regulates glucose transport in the rat L6 muscle cell line
Biochimica et Biophysica Acta 1344 Ž1997. 221–229
Acylation-stimulating protein ž ASP/ regulates glucose transport in the rat L6 muscle cell line Yuzhen Tao b, Katherine Cianflone a
a,)
, Allan D. Sniderman a , Susan P. Colby-Germinario b, Ralph J. Germinario b
McGill Unit for the PreÕention of CardioÕascular Disease, Royal Victoria Hospital, DiÕision of Cardiology, 687 Pine AÕenue West, Montreal, Que. H3A 1A1, Canada b Lady DaÕis Institute, Sir Mortimer B. DaÕis, Jewish General Hospital, McGill UniÕersity, Montreal, Que. H3T 1E2, Canada Received 25 July 1996; accepted 16 September 1996
Abstract Acylation-stimulating protein ŽASP., a human plasma protein, is a potent stimulator of triglyceride synthesis and glucose transport in both human adipocytes and fibroblasts. The purpose of the present in vitro study was to examine the effect of ASP on glucose transport in muscle cells. ASP stimulated 2-deoxy-glucose transport Ž2-DG. in differentiated rat L6 myotubes in a time Ž30 min to 24 h. and concentration dependent manner Ž97% increase.. The magnitude of the ASP effect on glucose transport was comparable to the time- and concentration-dependent effects seen with insulin Ž125% increase., but was additive to insulin, pointing to involvement of differential signalling pathways. ASP stimulation was dependent on cell differentiation in that glucose transport increased by only 12% in myoblasts, comparable to the effect of insulin in myoblasts Ž15% increase. demonstrating selective responsiveness of the differentiated myotubes to ASP and insulin. The mechanism for the ASP induced increase in glucose transport was also examined. ASP increased the Vmax for 2-DG transport by 183% Ž4.02 vs. 1.42 nmolrmg cell proteinr30 s; ASP vs. Control, respectively.. This could be explained by an increased translocation of glucose transporters ŽGLUT 1, GLUT 4 and GLUT 3. to the plasma membrane surface as demonstrated by Western analysis Žq43% P - 0.05, q30% P - 0.05, and q49% P - 0.05, respectively.. The effects of ASP were equal to those of insulin Žq47%, q26% and q53% for GLUT 1, GLUT 4 and GLUT 3, respectively. and in all cases were paralleled by comparable glucose transport increases under the same incubation conditions. After long-term stimulation Ž24 h., Western analysis indicated that ASP had a permissive effect on insulin stimulated increases in total GLUT3 and GLUT4 cellular transporter content. These results suggest that muscle is also responsive to ASP and that ASP may play a role in glucose metabolism in both muscle and adipose tissue. Keywords: Muscle; Glucose transport; Acylation stimulating protein; Complement 3A
1. Introduction Acylation stimulating protein ŽASP., the product of the interaction of three proteins — the third )
Corresponding author. Fax: q1 Ž514. 9820686.
component of complement ŽC3., factor B, and adipsin — is the most potent stimulant of triglyceride synthesis in human adipocytes yet described w1–3x. Production of the three precursor proteins is a differentiation dependent process in both human and murine adipocytes w2,4x and the amount of ASP produced by
0005-2760r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 0 0 0 5 - 2 7 6 0 Ž 9 6 . 0 0 1 4 4 - 0
222
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
the cells is proportional to the extent of differentiation w2,3x. Hence, ASP acts as a positive feedback stimulus for triglyceride synthesis in human adipocytes w5x. One of our principal objectives has been to elucidate the mechanism of action of ASP and a number of features have been established. In the first instance, specific interaction with the cell membrane appears essential and the extent of the biologic response of the cell — that is, the increase in triglyceride synthesis induced by ASP — is a function of the degree of this interaction w6x. In addition, the protein kinase C pathway has been demonstrated to be involved in the signal transduction mechanism by which ASP activates the appropriate intracellular metabolic responses w7x. Studies in cultured human skin fibroblasts established that ASP caused a marked increase in the rate of triglyceride synthesis by affecting two different processes of this synthetic pathway w8x. That is, ASP caused a significant increase in the activity of diacylglycerol acyltransferase, the last, and, in the case of adipose tissue, the rate-limiting enzyme involved in the biosynthesis of a triglyceride molecule, by increasing the Vmax of the enzyme w8,9x. ASP also caused an increase in specific membrane transport of glucose as manifested by increases in both 2-deoxyD-glucose Ž 2-DG. and 3-O-methyl-D-glucose transport w8x. Although the ASP effect was rapid in onset, it was not sustained after exposure of the cells to ASP for more than 6 h. The increase was concentration dependent, equivalent in magnitude to that of insulin, but independent and additive to insulin. ASP achieved this effect just as insulin does: by producing translocation of glucose transporters — Glut 1, in the case of these fibroblasts — from the intracellular vesicular fraction to the cell membrane w8x. The present studies were designed to examine the effect of ASP on specific glucose transport in myocytes, a physiologically important cell with respect to glucose transport. To do so, the effect of ASP on glucose transport in L6 rat myotubes was determined. The data demonstrate that ASP markedly increased specific membrane transport of glucose, that the effect was independent but additive to that of insulin, and that ASP produced translocation of GLUT 1, 3, and 4 glucose transporters. In contrast to fibroblasts, however, the effect of ASP on glucose transport was sustained for up to 24 h of ASP exposure.
2. Materials and methods 2.1. Materials 3
H-2-deoxy-D-glucose Ž2-DG. Žspecific activity, 34.2 Cirmmol and 125 I-labelled protein A were purchased from ICN Biochemicals Canada ŽMississauga, Ontario.. Insulin, bovine serum albumin ŽBSA., sucrose and HEPES were purchased from Sigma. Rabbit anti-Glut 4 antiserum was purchased from East Acres Biologicals ŽSouthbridge, MA.. Rabbit antiGlut 1 antiserum was prepared against purified human erythrocyte Glut 1 w10x and specifically recognizes the mouse Glut 1 transporter as well as the human Glut 1 w11x. Anti mouse Glut 3 rabbit polyclonal antisera specific for the C-terminal peptide sequences was a gift from the laboratory of Dr. Samuel Cushman ŽNIHrNIDDK, Bethesda, MD.. 2.2. Tissue culture L6 myoblasts Žobtained from Dr. Amira Klip, Toronto, Ontario. were grown in Dulbecco’s minimum essential medium ŽDMEM. containing 10% fetal bovine serum Ž FBS. Žvrv.. At approx. 75% confluence, before any myotube differentiation, myoblast studies were performed. For differentiation induction, 10% FBS-DMEM was removed and replaced with an equal volume of fusion medium ŽMEM containing 2% fetal bovine serum Žvrv... All experiments were done 4–5 days later when ) 85% of the cells were differentiated as assessed visually through morphology changes and myotube formation. Both 10% FBS-DMEM and 2% FBS-MEM were supplemented with a 100 = penicillinrstreptomycin antibiotic solution Ž10 mgrml penicillin and 10 mgrml streptomycin.. L6 cells were subcultured every 3 days at about 75% confluence using a split ratio of 1:2 after a 5 min incubation with versene followed by a 1 min incubation with 0.06% trypsin to detach cells from the flask w12x. 2.3. Glucose transport studies Myoblasts or myotubes grown in 35 mm diameter plastic Petri dishes ŽFalcon-Becton Dickinson Canada, Mississauga, Ontario. were employed. The cell monolayers were rinsed once in serum-free DMEM
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
containing 1 mgrml BSA and incubated overnight Ži.e., 18–24 h. in serum-free DMEM before exposure to the experimental conditions. In all experiments, when measuring glucose transport, zero-time controls were subtracted. Glucose transport was assessed by measuring the uptake of 2-deoxy-glucose Ž2-DG. . At the time of assay for glucose transport, the cell monolayers were rinsed two times with 2 ml glucose-free phosphate buffered saline Ž PBS. Ž pH 7.4. at 378C. Then, 0.8 ml PBS containing 3 H-2-DG Ž0.05 mM; specific activity 0.023 m Cirnmol. was added to each plate for the appropriate time interval Žusually 5 min. . After incubation, the radioactive medium was aspirated and the cell monolayers were washed 4 times Ž2 ml each time. with ice-cold PBS containing 5 m M Cytochalasin B. The incubation time used for the uptake of 2-DG was on the linear portion of the uptake curve w13,14x. The monolayers were dissolved in 1.0 N NaOH and aliquots taken for liquid scintillation counting and protein determination w15x. 2.4. ASP purification General chemicals and solvents were from Fisher Scientific Ž Nepean, Canada. . Acylation Stimulating Protein Ž ASP. was isolated from outdated frozen plasma obtained from the blood bank based on a modification w7x of a previous method w1x. Frozen plasma was thawed overnight at 48C, warmed quickly with stirring to 378C and spiked with inulin Ž 0.5% wrv final concentration. and with 1 M MgCl 2 solution Ž2 mM final concentration.. The spiked plasma was stirred for 1 h at 378C to enzymatically maximize in vitro generation of ASP from endogenous complement C3 w16x and increase the final yield of ASP. The plasma was transferred to an ice bath and globular proteins were precipitated by addition of concentrated HCl Ž1 M final concentration.. The plasma was centrifuged at 4000 g for 30 min and the supernatant was neutralized to pH 7.4 by the addition of 10 N NaOH. A C-18 Sep-Pak column ŽWatersMillipore, Mississauga, Ontario., 40 gr2 l of plasma was pre-equilibrated with 5 volumes of 80% ACNr0.1% TFA Žacetonitriler0.1% trifluoroacetic acid vrv. followed by 10 volumes of 0.1% TFA prior to use. The plasma supernatant was then applied
223
to the column and the column was washed with 10 volumes Ž until OD 280 returned to baseline. of 20% ACNr0.1% TFA. A single protein peak was eluted from the column with 10 volumes of 80% ACNr0.1% TFA and collected as 10 ml fractions ŽOD 280 nm.. The elution peak Ž average 60 ml. was pooled and loaded on an S-Sepharose column ŽPharmacia LKB Biotechnology Products, Baie d’Urfe, Quebec, Canada. 20 mlrbed volumerl starting plasma which had been previously pre-equilibrated with 5 volumes of buffer A Ž10 mM Tris, 10 mM NaCl, pH:7.1.. After loading the sample, the column was washed with 5 volumes of buffer A Žuntil OD 280 returned to baseline. and eluted with 10 volumes of 1M NaCl in buffer A. The elution peak Žaverage 100 ml. was pooled and loaded on a semi-preparative Vydac Protein C4 column Ž1.0 = 25 cm, Separations Group, Hesperia, CA. pre-equilibrated with 0.1% TFA Ž5 runsr2 l starting plasma.. Loading and elution was carried out at 2.0 mlrmin with OD monitoring at 280 nm. Following a 20 min wash of the column with 24% ACNr0.1% TFA, proteins were eluted from the column and fractionated Ž3 mlrtube. with a linear gradient from 24% ACNr0.1% TFA to 80% ACNr0.1% TFA in 45 min. The ability of column fractions to stimulate triglyceride synthesis was tested as previously described w1x. Tubes from the peak containing triglyceride synthesis stimulating activity from sequential runs were pooled. Following this protocol, ASP typically eluted from the Vydac C4 column at 50% ACN. The pool was aliquoted into 20 siliconized glass vials ŽSigmacote, Sigma, St. Louis, MO see supplier’s instructions. and lyophilized in a centrifuge evaporator. Lyophilized aliquots were reconstituted in 1 ml sterile PBS with gentle vortexing and stored at y808C. It should be noted that, ASP should be stored at y808C in siliconized glass vials to reduce aggregation, sticking to surfaces and other non-specific losses. Average yield from 1 l plasma was 19 " 3.3 mg as measured by commercial radioimmunoassay for C3adesArg ŽAmersham, Oakville, Canada. . 2.5. Membrane isolation Myotubes were induced as described above. A crude plasma membrane fraction was prepared as
224
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
described previously w17,18x. Briefly, the cells from a 100 mm diameter plastic Petri plate were scraped and centrifuged Ž700 = g . for 10 min and placed on ice. The cell pellets were resuspended in buffer containing 250 mM sucrose, 5 mM NaN3 , 2 mM EGTA, 100 mM phenylmethylsulfonylfluoride, 1 m M leupeptin, 1 m M pepstatin and 20 mM Hepes ŽpH 7.4. and homogenized using 20 strokes of a Dounce homogenizer. The homogenate was centrifuged at 760 = g for 5 min after which the supernatant was centrifuged at 31 000 = g for 60 min. The pellet Ži.e., the crude plasma membrane. was resuspended in the homogenization buffer stored at y708C and employed for western analysis. Plasma membrane markers were assayed to verify plasma membrane enrichment. Ouabain sensitive Naq-ATPase w19x activities for the homogenate were 16 " 9.2, 12 " 3.2 and 15 " 3.2 for control, insulin-treated and ASP-treated cells, respectively. For the crude plasma membranes, activities were 57 " 12, 57.6 " 19 and 54.7 " 16 for control, insulin-treated and ASP-treated cells, respectively, measured as phosphatermg protein per 60 min Ž n s 2 experiments. which ranges from 3- to 4.8-fold enrichment in the crude plasma membrane fraction. 2.6. SDS gel electrophoresis and western blots Purified human erythrocyte glucose transporter was prepared as described elsewhere w10x. A crude extract of rat brain cortex was employed as a standard for Glut 3 Ža gift from the laboratory of Dr. S. Cushman.. Western blot analysis was carried out as follows: 1. samples were solubilized in Laemmli w20x sample buffer and electrophoresis was performed on 10% polyacrylamide slab gels as described earlier w17,18x. 2. Proteins were blotted onto nitrocellulose paper Ž 0.2 m m; Schleicher and Schullx that was blocked with 1% BSA Ž 1 h at 208C.. 3. This was followed by incubation with specific antibody Žovernight at 48C.. 4. Subsequently, 125 I-labelled protein A at 1.2 = 10 6 dpmr10 ml in blocking buffer containing 0.1% Triton X-100 was added for 1 h at 208C. 5. Autoradiography was performed on Kodak XAR film ŽEastman Kodak, Rochester, NY. exposed at y858C overnight. Spots on the autoradiograms were marked using the developed X-ray film, cut out and counted in a gamma counter for quantitation.
2.7. Statistics All results are expressed as average" S.E.M. Student’s t-test Žtwo-tailed. for two means was used as a comparison between two different treatment groups. A one-way ANOVA was used when comparing more than two means.
3. Results The data in Fig. 1 demonstrate the effect of increasing insulin exposure ŽFig. 1, top. and increasing insulin concentration ŽFig. 1, bottom. on differentiated L6 myotubes. The effect of insulin on 2-DG transport was maximal after 1 h exposure at an insulin concentration of 0.067 m M ŽFig. 1, top. where basal glucose transport is 249 " 54 pmol 2-DGrmg protein per 5 min. These data are similar to those previously reported by Klip and colleagues employing the same cell type and mode of induction of differentiation w17,18x. As shown in Fig. 1 bottom, this effect of insulin was concentration dependent. Having established that this muscle cell model system was behaving as expected, we sought to determine whether these cells would respond to ASP. The data in Fig. 2 top, clearly indicate that L6 myotubes respond very well to ASP Ž 2.8 m M. exhibiting an increase in 2-DG transport after 30 min exposure which increased up to 1–2 h. In fact, the ASP response curve was nearly identical to the insulin response curve over the 2 h time frame of the experiment as also shown in Fig. 2. Of considerable interest was the observation that the combined effects of ASP Ž2.8 m M. plus insulin Ž0.067 m M. exhibited an increase in 2-DG transport well above the individual treatments suggesting that the responses were additive. This ASP response by the L6 cells was dose dependent with the greatest response to ASP occurring at an ASP concentration of 2.8 m M ŽFig. 2, bottom. Ž half maximal concentrations 0.5 m M.. As shown in Table 1, ASP had no effect on the K m of 2-DG transport in myotubes Ž1.1 mM vs. 1.8 mM in the control vs. ASP groups respectively. while the Vmax for 2-DG transport was increased by 183% Ž4.02 vs. 1.42 nmolrmg protein per 30 s; ASP vs. Control respectively. . We then determined if the response to ASP of
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
225
undifferentiated myoblasts and differentiated myotubes differed. The data in Table 2 compare the responses of myoblasts and myotubes to insulin, ASP, and a combination of the two factors. The myoblasts exhibited marginal increases in glucose transport to all factors Ža range of 12% to 15% increase above
Fig. 1. Time- and concentration-dependent effect of insulin on glucose transport in L6 myotubes. Myoblasts were grown to 75% confluence, then differentiated over 4–5 days to myotubes. Top panel: myotubes were exposed to 0.067 m M insulin for 15 min to 2 h and 2-deoxy-glucose Ž2-DG. transport was then assessed. Results are expressed as 2-DG Uptake Ratio calculated as Insulin treatedrcontrol for an average of ns 2 experiments Žtriplicate plates in all experiments. where basal glucose transports 249" 54 pmol 2-DGrmg protein per 5 min. Bottom panel: myotubes were exposed for 1 h to increasing concentrations of insulin as indicated followed by assessment of glucose transport. Results are shown for a representative experiment Žtriplicate plates for each point. and are expressed as 2-DG Uptake Ratio calculated as Insulin treatedrcontrol where basal glucose transports 586" 33 pmol 2-DGrmg protein per 5 min.
Fig. 2. Time- and concentration-dependent effect of ASP on glucose transport in L6 myotubes. Myoblasts were differentiated as in Fig. 1. Top panel: myotubes were exposed to 2.8 m M ASP Žv . or 0.067 m M insulin Ž ) . Žor both together ŽB. for 30 min to 2 h and 2-deoxy-glucose Ž2-DG. transport was then assessed and compared to Control ŽP.. Results are expressed as pmoles 2DGrmg cell protein ŽP. per 5 min for an average of ns 3 experiments Žtriplicate plates in each experiment.. Bottom panel: myotubes were exposed for 1 h to increasing concentrations of ASP followed by assessment of glucose transport. Results are expressed as pmoles 2-DGrmg cell protein ŽP. per 5 min for an average of ns 4 experiments; P - 0.05 by one-way ANOVA.
control. where basal levels of glucose transport were 2.279 nmol 2-DGrmg cell protein, 5 min for myoblasts and 4.222 for myotubes. The responses of the myotubes, however, were significantly elevated with all treatments versus the myoblasts Ža range of 44% to 77% increase above control; P - 0.01 by one-way ANOVA. . Since 2-DG transport was increased in response to ASP, we sought to determine the precise
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
226
Table 1 The effect of ASP on the K m and Vmax of 2-deoxy-glucose Ž2-DG. Treatment
2-DG transport
Ž1 h.
K m ŽmM.
y ASP q ASP Ž2.8 m M.
1.1"0.19 1.8"0.16
Vmax Žnmol 2-DGr mgP per 30 s. )
1.42"0.21 4.02"0.56
))
)
No significant difference P ) 0.05. Significant difference P - 0.05. Cells were differentiated as described in Section 2, serum deprived overnight and exposed to ASP Ž2.8 m M. for 1 h. Seven different 2-DG concentrations were employed Ž0.05 mM to 5.0 mM.. The uptake time was 30 s Žlinear for all concentrations. and saturable sugar uptake was determined by subtracting the diffusional uptake using L-glucose w3x. Data is mean"S.E.M. and represents the average of 2 experiments, with duplicate plates at all concentrations. ))
mechanismŽ s. that were involved in this ASP induced increase in glucose transport. L6 myotubes were exposed to insulin Ž 0.067 m M. or ASP Ž2.8 m M. for 1 h and a crude plasma membrane preparation was isolated using identical procedures as described elsewhere w17,18x. Plasma membrane extracts were then separated by molecular weight on SDS polyacrylamide gel electrophoresis and probed using anti-glucose transporter antibodies. The data in Fig. 3 Žpanel A1, A2, A3. show the results of the Western blot Table 2 The effect of ASP and insulin on 2-deoxy-glucose Ž2-DG. transport ratios in myoblasts and myotubes Treatment
2-DG transport ratios
Ž1 h.
myoblasts
myotubes
ASP Ž2.8 m M. ASP Ž2.8 m M.qinsulin Ž0.067 m M. Insulin Ž0.067 m M.
1.12"0.04 1.14"0.02
1.55"0.08 1.70"0.08
)
1.15"0.01
1.47"0.06
)
)
)
One-way ANOVA indicated that all transport ratios were significantly different from those observed for the myoblasts Ž F s 19.53; P - 0.01.. Myoblasts were grown to near confluence Ž75%. and myotubes were induced in 2% FBS MEM for 5 days as described in Section 2. 2-DG transport was determined after 18 h in serum-free DMEM Ž0.05 mM 2-DG for minutes.. All treatments were for 1 h. Data represent the average of 6 experiments"S.E.M. where basal glucose transport is 2.279 Žmyoblasts. and 4.222 Žmyotubes. nmolrmg protein per 5 min.
Fig. 3. Western analysis of ASP and insulin effects on GLUT 1, GLUT 3, and GLUT 4 glucose transporters in L6 myotube plasma membrane preparations. Myoblasts were differentiated as in Fig. 1. L6 myotubes were exposed to ASP Ž2.8 m M. or insulin Ž0.067 m M. for 1 h and a crude plasma membrane was then isolated. Western analysis was performed and the autoradiograms for GLUT 1 Žpanel A1., GLUT 3 Žpanel A2. and GLUT 4 Žpanel A3. are shown. The bands on the Western blots were cut out and counted in a gamma counter for quantitation Žpanels B1, B2 and B3 represent the GLUT1, GLUT3 and GLUT4 Western blots, respectively. Lanes a, b and c in all panels represent ASP, insulin and control respectively. Results are expressed as 125 I protein A Žcpm. for an average of 5 to 6 experiments each"S.E.M., ) P - 0.05 vs. control.
analysis of GLUT1, GLUT3 and GLUT4 transporters in this plasma membrane preparation. In Fig. 3 Žpanel B1. there was a clear increase in the amount of GLUT 1 transporter in the membranes in response to stimulation by insulin Žlane b, 47% average increase, P - 0.05. and, importantly, in response to ASP Žlane a, 43% average increase, P - 0.05.. A predominant GLUT 1 doublet was apparent in the ASP treated group but was not noticeable in any other treatment group. In this same series of experiments, the membrane preparations were probed for the GLUT 4 transporter ŽFig. 3, panel A3. . The data indicate that both insulin Žlane b. and ASP Žlane a. significantly increased the GLUT 4 transporter content of the membrane preparations by an average 26% and 30%, respectively Ž P - 0.05 for both ASP and insulin vs. control.. No doublets of the GLUT 4 transporter were observed in any experiment. Continued analysis was performed on the same membrane preparations for
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
the presence of the GLUT 3 transporter ŽFig. 3, panel A2.. The data in Fig. 3 Žpanel B2., show that ASP and insulin Žlanes a and b respectively. increased the GLUT 3 transporter content of the membranes by an average 49% Ž P - 0.05. and 53% Ž P - 0.05., respectively. Again, no doublets of this transporter were observed in any experiments. The stimulation of glucose transport done in parallel with the translocation experiments yielded a 67% average increase in the insulin group and a 40% average increase in the ASP group Žboth treatments were significantly different from the control group, two-tailed t-test, P - 0.01, n s 6.. Having observed short-term effects of ASP on 2-DG transport, we next investigated the effects of long-term exposure Žup 24 h. by ASP, insulin or a combination thereof on 2-DG transport. Over the 24 h period in serum-free medium basal glucose transport decreases slightly although not significantly. This is probably due to the extended serum-free incubation in a high concentration of glucose. The data seen in Table 3 clearly demonstrate that the effects of ASP on 2-DG transport are present from 2–24 h Ž 38% increase after 2 h, 96% after 8 h, 78% after 8 h and 97% after 24 h.. In addition, the effects of insulin were also elevated over these same time frames Ž45%; 93%; 106% and 125%, respectively.. Further, the effects of the combined treatments were significantly elevated over either treatment group suggesting a continued enhancement of 2-DG transport over long time frames Ž 96%; 160%; 199% and 232%; increase over basal at 2–24 h, respectively.. The total cellular content of the various glucose transporters was determined after 24 h exposure to the different treatments ŽFig. 4.. The GLUT1 trans-
227
Fig. 4. Western analysis of ASP and insulin effects on total cellular GLUT1, GLUT3 and GLUT4 glucose transporters in L6 myotubes. L6 myotubes were differentiated as described in Section 2, incubated overnight in serum-free medium and then incubated in basal medium Žopen bars., with ASP Ž2.8 m M, solid bars., insulin Ž0.067 m M, striped bars. or both Žcross-hatched bars. for 24 h. Cells were harvested and total glucose transporters assessed by Western analysis. Results are presented for an average of 3 experiments"S.E.M., where ) s P - 0.05 versus control and as P - 0.05 for ASPqinsulin versus ASP alone.
porter showed a trend towards increased content following treatment with ASP, however, the differences were not significant. ASP alone had no effect on total cellular content of GLUT3 or GLUT4 although ASP did increase translocation of GLUT3 and GLUT4 ŽFig. 3. and overall glucose transport. However, ASP did enhance significantly the insulin-stimulated increase on total cellular transporter content for both GLUT3 Ž166% ASP q insulin versus 143% insulin alone, P - 0.05 where basal s 100%. and GLUT4 Ž276% ASP q insulin versus 205% insulin alone, P - 0.05 where basal s 100%.. Thus, ASP demonstrates both short-term and
Table 3 The effect of long-term exposure to ASP and insulin in L6 myotubes2-DG transport Žpmolrmg protein per 5 min. Time Žh. 2 4 8 24 a
Control 852 " 36 804 " 39 708 " 66 607 " 84
qASP Ž2.8 m M. 1180 " 138 1578 " 440 1264 " 66 1201 " 146
a a
qInsulin Ž.067 m M. a
1238 " 80 1545 " 103 1453 " 164 1374 " 186
a a
ASP q insulin Ž2.8 m M q .067 m M. 1667 " 137 a,b 2091 " 489 a,b 2120 " 314 a,b 2021 " 371 a,b
Significant difference vs. control Ž P - 0.05.. Significant difference vs. ASP- or insulin-treated P groups Ž P - 0.05.. Cells were treated as described in Section 2 for the times indicated above. 2-DG transport was assayed at a concentration of 0.05 mM 2-DG for 5 min. Data represent the average of three experiments Žtriplicate plates in each experiment. " S.E.M. Statistical analysis was a one-way repeated measures ANOVA Ž F s 35.87 at 2 h; 23.15 at 4 h; 19.91 at 8 h and 11.5 at 24 h..
b
228
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
long-term effects on glucose transport via effects on total cellular glucose transport content and translocation of glucose transporters to the plasma membrane.
4. Discussion The present studies confirm previous data as to the effect of insulin on glucose transport in L6 myotubes w17,18,21x. Importantly the data also demonstrate that ASP markedly stimulates specific membrane transport of glucose in L6 myotubes, an effect which occurred at physiological concentrations of ASP. As with cultured human skin fibroblasts, the effect of ASP on this process was concentration and time dependent and was comparable to that of insulin. However, the effect of ASP was independent of insulin and additive to insulin. Given the evidence that suggests that different activation cascades are employed by ASP and insulin, this independence and additivity of effects should not be surprising w7x. The effect of both ASP and insulin on glucose transport was pronounced in the differentiated myotubes, but there was little effect in the undifferentiated myoblasts pointing to a selective differentiation induced response to both factors. Interestingly, we have previously shown that differentiated human adipocytes are more responsive than undifferentiated human preadipocytes with respect to ASP stimulation of triglyceride synthesis w2x. Experiments were carried out to determine the specific mechanism by which ASP caused the membrane transport of glucose to increase. Our data clearly demonstrate that ASP treatment resulted in a significant increases in the myotube plasma membrane content of GLUT1, GLUT3 and GLUT4 glucose transporters indicating that translocation of all three glucose transporters was effected by exposure to physiological levels of ASP. Insulin stimulation of the GLUT1, GLUT3 and GLUT4 glucose transporters was 47%, 64% and 26% greater than control membranes while ASP stimulation of the same transporters, respectively, was 43%, 62% and 30% greater than control membranes. The extent of translocation paralleled 2-DG uptake in parallel cultures treated with insulin Ža 67% average increase. or ASP Ža 40% average increase.. The protein kinase C pathway has been demonstrated to be involved in the signal trans-
duction mechanism by which ASP activates the appropriate intracellular metabolic responses Ž triglyceride synthesis and glucose transport. w7x and may be involved in translocation of glucose transporters. Long-term exposure of myotubes to ASP Ži.e., for up to 24 h. resulted in continued elevated rates of glucose transport ŽTable 3.. These results suggest that ASP as well as insulin can effect long-term increases in glucose transport in muscle. Our previous observations on glucose transport after long-term exposure of human fibroblasts to ASP w8x contrast with the present effects of ASP on muscle cell cultures. In those studies, ASP stimulation of glucose transport peaked 2 h post ASP exposure, was decreasing by 8 h and, after 24 h exposure to ASP, glucose transport rates were no longer increased. In the present study, glucose transport is elevated up to 24 h. Of considerable interest is that the ASP q insulin group is significantly elevated at 8 and 24 h post treatment when compared to ASP or insulin alone. The data in Fig. 4 which reflects the total amount of cellular glucose transporters present after 24 h treatment indicate that the elevations in transport observed are associated with increased cellular levels of GLUT3 and GLUT4. Thus, the differences observed most likely reflect the differences between the effect of ASP on the undifferentiated fibroblast vs. the differentiated state of L6 myotubes. That the biologic effect persists over prolonged periods with the differentiated cell further points to physiologic relevance of the observations. Our previous work has focused on the role of ASP in human triglyceride metabolism. Thus, following an oral fat load, but not after an oral glucose load, plasma ASP levels gradually rise and fall, roughly paralleling the changes in plasma triglycerides w22x. In normals, the rate at which triglycerides were cleared was related to both the fasting and peak plasma levels of ASP consistent with the hypothesis that plasma ASP is a determinant of peripheral triglyceride clearance. ASP stimulates triglyceride synthesis markedly in differentiated human adipocytes and preliminary results indicate that this is also true of glucose transport in differentiated human adipocytes w23x. Adipocytes from patients with hyperapoB synthesize triglycerides less rapidly than normal w24x and fibroblasts from such patients respond
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
less well to ASP w6x, suggesting that dysfunction of the adipsin-ASP pathway may be important in the pathogenesis of this dyslipoproteinemia. By contrast, in patients with gynoid obesity, adipocytes remain responsive to ASP w25x and plasma levels of ASP are markedly elevated in such patients w26,27x. Studies in cultured cells indicate that ASP acts via a second messenger system involving protein kinase C w7x. This cell triggering is likely mediated through specific interaction with a cell surface receptor w6x. Studies to identify and characterize the ASP receptor are ongoing. Given the similarities which have been demonstrated between the mechanisms responsible for carrier-mediated glucose transport in human muscle cells in culture and L6 myotubes w28x, the present data raise the possibility that ASP may also act in vivo on muscle, a traditional insulin target tissue, as well as adipocytes. If this is the case, then ASP may play a broader role in glucose transport than previously recognized.
Acknowledgements This research was supported by a grant from the Medical Research Council of Canada to Dr. K. Cianflone. Dr. K. Cianflone is the recipient of a scholarship of the Heart and Stroke Foundation of Canada and ‘les Fonds de la Recherche en Sante´ du Quebec’. ´
References w1x Baldo, A., Sniderman, A.D., St-Luce, S., Avramoglu, R.K., Maslowska, M., Hoang, B., Monge, J.C., Bell, A., Mulay, S. and Cianflone, K. Ž1993. J. Clin. Invest. 92Ž3., 1543– 1557. w2x Cianflone, K., Roncari, D.A.K., Maslowska, M., Baldo, A., Forden, J. and Sniderman, A.D. Ž1994. Biochemistry 33, 9489–9495. w3x Cianflone, K., and Maslowska, M. Ž1995. Eur. J. Clin. Invest. 25, 817–825. w4x Choy, L.N., Rosen, B.S. and Spiegelman, B.M. Ž1992. J. Biol. Chem. 367, 12736–12741.
229
w5x Sniderman, A.D. and Cianflone, K. Ž1995. in Yearbook of Endocrinology, pp. xix–xxxvii, Mosby, Chicago, IL. w6x Cianflone, K., Maslowska, M. and Sniderman, A.D. Ž1990. J. Clin. Invest. 85Ž3., 722–730. w7x Baldo, A., Sniderman, A.D., St-Luce, S., Zhang, X.J. and Cianflone, K. Ž1995. J. Lipid Res. 36, 1415–1426. w8x Germinario, R., Sniderman, A.D., Manuel, S., Pratt, S., Baldo, A. and Cianflone, K. Ž1993. Metabolism 40Ž5., 574–580. w9x Yasruel, Z., Cianflone, K., Sniderman, A.D., Rosenbloom, M., Walsh, M. and Rodriguez, M.A. Ž1991. Lipids 26Ž7., 495–499. w10x Baldwin, S.A., Baldwin, J.M. and Lienhard, G.E. Ž1982. Biochemistry 21, 3836–3842. w11x Pratt, S.E., Colby-Germinario, S.P., Manuel, S. and Germinario, R.J. Ž1994. J. Cell Physiol. 161, 580–588. w12x Germinario, R.J., Rockman, H., Olveira, M. and Manuel, S. Ž1982. J. Cell Physiol. 112, 367–372. w13x Germinario, R.J., Lakshmi, T.M. and Thirion, J.P. Ž1989. J. Cell Physiol. 138, 300–304. w14x Germinario, R.J., Kristof, A., Chang, Z. and Manuel, S. Ž1990. J. Cell Physiol. 145, 318–323. w15x Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. Ž1951. J. Biol. Chem. 193, 265–275. w16x Hugli, T.E. Ž1975. J. Biol. Chem. 250, 8293–8301. w17x Mitsumoto, Y., Burdett, E., Grant, A. and Klip, A. Ž1991. Biochem Biophys. Res. Commun. 175, 652–659. w18x Bilan, P., Mitsumoto, Y., Maher, F., Simpson, I. and Klip, A. Ž1992. Biochem. Biophys. Res. Commun. 186, 1129– 1137. w19x Hildago, C., Gonzales, M.E., and Lagos, R. Ž1983. Arch. Biochem. Biophys. 253, 279–286. w20x Laemmli, U.K. Ž1970. Nature 227, 680–685. w21x Mitsumoto, Y. and Klip, A. Ž1992. J. Biol. Chem. 267, 4957–4962. w22x Cianflone, K., Vu, H., Walsh, M., Baldo, A. and Sniderman, A.D. Ž1989. J. Lipid Res. 30Ž11., 1727–1733. w23x Maslowska, M., Sniderman, A.D., Germinario, R. and Cianflone, K. Ž1995. Clin. Invest. Med. 18ŽSuppl. B., B42. w24x Teng, B., Forse, A., Rodriguez, M.A. and Sniderman, A.D. Ž1988. Can. J. Pharmacol. Physiol. 66Ž3., 239–242. w25x Walsh, M.J., Sniderman, A.D., Cianflone, K., Vu, H., Rodriguez, M.A. and Forse, R.A. Ž1989. J. Surg. Res. 46, 470–473. w26x Sniderman, A.D., Cianflone, K. and Eckel, R.H. Ž1991. Int. J. Obes. 15Ž5., 333–336. w27x Cianflone, K., Kalant, D., Marliss, E.B., Gougeon, R. and Sniderman, A.D. Ž1995. Int. J. Obes. 19, 604–609. w28x Sarabia, V., Ramlal, T. and Klip, A. Ž1990. Biochem. Cell Biol. 68, 536–542.
Acylation-stimulating protein ž ASP/ regulates glucose transport in the rat L6 muscle cell line Yuzhen Tao b, Katherine Cianflone a
a,)
, Allan D. Sniderman a , Susan P. Colby-Germinario b, Ralph J. Germinario b
McGill Unit for the PreÕention of CardioÕascular Disease, Royal Victoria Hospital, DiÕision of Cardiology, 687 Pine AÕenue West, Montreal, Que. H3A 1A1, Canada b Lady DaÕis Institute, Sir Mortimer B. DaÕis, Jewish General Hospital, McGill UniÕersity, Montreal, Que. H3T 1E2, Canada Received 25 July 1996; accepted 16 September 1996
Abstract Acylation-stimulating protein ŽASP., a human plasma protein, is a potent stimulator of triglyceride synthesis and glucose transport in both human adipocytes and fibroblasts. The purpose of the present in vitro study was to examine the effect of ASP on glucose transport in muscle cells. ASP stimulated 2-deoxy-glucose transport Ž2-DG. in differentiated rat L6 myotubes in a time Ž30 min to 24 h. and concentration dependent manner Ž97% increase.. The magnitude of the ASP effect on glucose transport was comparable to the time- and concentration-dependent effects seen with insulin Ž125% increase., but was additive to insulin, pointing to involvement of differential signalling pathways. ASP stimulation was dependent on cell differentiation in that glucose transport increased by only 12% in myoblasts, comparable to the effect of insulin in myoblasts Ž15% increase. demonstrating selective responsiveness of the differentiated myotubes to ASP and insulin. The mechanism for the ASP induced increase in glucose transport was also examined. ASP increased the Vmax for 2-DG transport by 183% Ž4.02 vs. 1.42 nmolrmg cell proteinr30 s; ASP vs. Control, respectively.. This could be explained by an increased translocation of glucose transporters ŽGLUT 1, GLUT 4 and GLUT 3. to the plasma membrane surface as demonstrated by Western analysis Žq43% P - 0.05, q30% P - 0.05, and q49% P - 0.05, respectively.. The effects of ASP were equal to those of insulin Žq47%, q26% and q53% for GLUT 1, GLUT 4 and GLUT 3, respectively. and in all cases were paralleled by comparable glucose transport increases under the same incubation conditions. After long-term stimulation Ž24 h., Western analysis indicated that ASP had a permissive effect on insulin stimulated increases in total GLUT3 and GLUT4 cellular transporter content. These results suggest that muscle is also responsive to ASP and that ASP may play a role in glucose metabolism in both muscle and adipose tissue. Keywords: Muscle; Glucose transport; Acylation stimulating protein; Complement 3A
1. Introduction Acylation stimulating protein ŽASP., the product of the interaction of three proteins — the third )
Corresponding author. Fax: q1 Ž514. 9820686.
component of complement ŽC3., factor B, and adipsin — is the most potent stimulant of triglyceride synthesis in human adipocytes yet described w1–3x. Production of the three precursor proteins is a differentiation dependent process in both human and murine adipocytes w2,4x and the amount of ASP produced by
0005-2760r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 0 0 0 5 - 2 7 6 0 Ž 9 6 . 0 0 1 4 4 - 0
222
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
the cells is proportional to the extent of differentiation w2,3x. Hence, ASP acts as a positive feedback stimulus for triglyceride synthesis in human adipocytes w5x. One of our principal objectives has been to elucidate the mechanism of action of ASP and a number of features have been established. In the first instance, specific interaction with the cell membrane appears essential and the extent of the biologic response of the cell — that is, the increase in triglyceride synthesis induced by ASP — is a function of the degree of this interaction w6x. In addition, the protein kinase C pathway has been demonstrated to be involved in the signal transduction mechanism by which ASP activates the appropriate intracellular metabolic responses w7x. Studies in cultured human skin fibroblasts established that ASP caused a marked increase in the rate of triglyceride synthesis by affecting two different processes of this synthetic pathway w8x. That is, ASP caused a significant increase in the activity of diacylglycerol acyltransferase, the last, and, in the case of adipose tissue, the rate-limiting enzyme involved in the biosynthesis of a triglyceride molecule, by increasing the Vmax of the enzyme w8,9x. ASP also caused an increase in specific membrane transport of glucose as manifested by increases in both 2-deoxyD-glucose Ž 2-DG. and 3-O-methyl-D-glucose transport w8x. Although the ASP effect was rapid in onset, it was not sustained after exposure of the cells to ASP for more than 6 h. The increase was concentration dependent, equivalent in magnitude to that of insulin, but independent and additive to insulin. ASP achieved this effect just as insulin does: by producing translocation of glucose transporters — Glut 1, in the case of these fibroblasts — from the intracellular vesicular fraction to the cell membrane w8x. The present studies were designed to examine the effect of ASP on specific glucose transport in myocytes, a physiologically important cell with respect to glucose transport. To do so, the effect of ASP on glucose transport in L6 rat myotubes was determined. The data demonstrate that ASP markedly increased specific membrane transport of glucose, that the effect was independent but additive to that of insulin, and that ASP produced translocation of GLUT 1, 3, and 4 glucose transporters. In contrast to fibroblasts, however, the effect of ASP on glucose transport was sustained for up to 24 h of ASP exposure.
2. Materials and methods 2.1. Materials 3
H-2-deoxy-D-glucose Ž2-DG. Žspecific activity, 34.2 Cirmmol and 125 I-labelled protein A were purchased from ICN Biochemicals Canada ŽMississauga, Ontario.. Insulin, bovine serum albumin ŽBSA., sucrose and HEPES were purchased from Sigma. Rabbit anti-Glut 4 antiserum was purchased from East Acres Biologicals ŽSouthbridge, MA.. Rabbit antiGlut 1 antiserum was prepared against purified human erythrocyte Glut 1 w10x and specifically recognizes the mouse Glut 1 transporter as well as the human Glut 1 w11x. Anti mouse Glut 3 rabbit polyclonal antisera specific for the C-terminal peptide sequences was a gift from the laboratory of Dr. Samuel Cushman ŽNIHrNIDDK, Bethesda, MD.. 2.2. Tissue culture L6 myoblasts Žobtained from Dr. Amira Klip, Toronto, Ontario. were grown in Dulbecco’s minimum essential medium ŽDMEM. containing 10% fetal bovine serum Ž FBS. Žvrv.. At approx. 75% confluence, before any myotube differentiation, myoblast studies were performed. For differentiation induction, 10% FBS-DMEM was removed and replaced with an equal volume of fusion medium ŽMEM containing 2% fetal bovine serum Žvrv... All experiments were done 4–5 days later when ) 85% of the cells were differentiated as assessed visually through morphology changes and myotube formation. Both 10% FBS-DMEM and 2% FBS-MEM were supplemented with a 100 = penicillinrstreptomycin antibiotic solution Ž10 mgrml penicillin and 10 mgrml streptomycin.. L6 cells were subcultured every 3 days at about 75% confluence using a split ratio of 1:2 after a 5 min incubation with versene followed by a 1 min incubation with 0.06% trypsin to detach cells from the flask w12x. 2.3. Glucose transport studies Myoblasts or myotubes grown in 35 mm diameter plastic Petri dishes ŽFalcon-Becton Dickinson Canada, Mississauga, Ontario. were employed. The cell monolayers were rinsed once in serum-free DMEM
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
containing 1 mgrml BSA and incubated overnight Ži.e., 18–24 h. in serum-free DMEM before exposure to the experimental conditions. In all experiments, when measuring glucose transport, zero-time controls were subtracted. Glucose transport was assessed by measuring the uptake of 2-deoxy-glucose Ž2-DG. . At the time of assay for glucose transport, the cell monolayers were rinsed two times with 2 ml glucose-free phosphate buffered saline Ž PBS. Ž pH 7.4. at 378C. Then, 0.8 ml PBS containing 3 H-2-DG Ž0.05 mM; specific activity 0.023 m Cirnmol. was added to each plate for the appropriate time interval Žusually 5 min. . After incubation, the radioactive medium was aspirated and the cell monolayers were washed 4 times Ž2 ml each time. with ice-cold PBS containing 5 m M Cytochalasin B. The incubation time used for the uptake of 2-DG was on the linear portion of the uptake curve w13,14x. The monolayers were dissolved in 1.0 N NaOH and aliquots taken for liquid scintillation counting and protein determination w15x. 2.4. ASP purification General chemicals and solvents were from Fisher Scientific Ž Nepean, Canada. . Acylation Stimulating Protein Ž ASP. was isolated from outdated frozen plasma obtained from the blood bank based on a modification w7x of a previous method w1x. Frozen plasma was thawed overnight at 48C, warmed quickly with stirring to 378C and spiked with inulin Ž 0.5% wrv final concentration. and with 1 M MgCl 2 solution Ž2 mM final concentration.. The spiked plasma was stirred for 1 h at 378C to enzymatically maximize in vitro generation of ASP from endogenous complement C3 w16x and increase the final yield of ASP. The plasma was transferred to an ice bath and globular proteins were precipitated by addition of concentrated HCl Ž1 M final concentration.. The plasma was centrifuged at 4000 g for 30 min and the supernatant was neutralized to pH 7.4 by the addition of 10 N NaOH. A C-18 Sep-Pak column ŽWatersMillipore, Mississauga, Ontario., 40 gr2 l of plasma was pre-equilibrated with 5 volumes of 80% ACNr0.1% TFA Žacetonitriler0.1% trifluoroacetic acid vrv. followed by 10 volumes of 0.1% TFA prior to use. The plasma supernatant was then applied
223
to the column and the column was washed with 10 volumes Ž until OD 280 returned to baseline. of 20% ACNr0.1% TFA. A single protein peak was eluted from the column with 10 volumes of 80% ACNr0.1% TFA and collected as 10 ml fractions ŽOD 280 nm.. The elution peak Ž average 60 ml. was pooled and loaded on an S-Sepharose column ŽPharmacia LKB Biotechnology Products, Baie d’Urfe, Quebec, Canada. 20 mlrbed volumerl starting plasma which had been previously pre-equilibrated with 5 volumes of buffer A Ž10 mM Tris, 10 mM NaCl, pH:7.1.. After loading the sample, the column was washed with 5 volumes of buffer A Žuntil OD 280 returned to baseline. and eluted with 10 volumes of 1M NaCl in buffer A. The elution peak Žaverage 100 ml. was pooled and loaded on a semi-preparative Vydac Protein C4 column Ž1.0 = 25 cm, Separations Group, Hesperia, CA. pre-equilibrated with 0.1% TFA Ž5 runsr2 l starting plasma.. Loading and elution was carried out at 2.0 mlrmin with OD monitoring at 280 nm. Following a 20 min wash of the column with 24% ACNr0.1% TFA, proteins were eluted from the column and fractionated Ž3 mlrtube. with a linear gradient from 24% ACNr0.1% TFA to 80% ACNr0.1% TFA in 45 min. The ability of column fractions to stimulate triglyceride synthesis was tested as previously described w1x. Tubes from the peak containing triglyceride synthesis stimulating activity from sequential runs were pooled. Following this protocol, ASP typically eluted from the Vydac C4 column at 50% ACN. The pool was aliquoted into 20 siliconized glass vials ŽSigmacote, Sigma, St. Louis, MO see supplier’s instructions. and lyophilized in a centrifuge evaporator. Lyophilized aliquots were reconstituted in 1 ml sterile PBS with gentle vortexing and stored at y808C. It should be noted that, ASP should be stored at y808C in siliconized glass vials to reduce aggregation, sticking to surfaces and other non-specific losses. Average yield from 1 l plasma was 19 " 3.3 mg as measured by commercial radioimmunoassay for C3adesArg ŽAmersham, Oakville, Canada. . 2.5. Membrane isolation Myotubes were induced as described above. A crude plasma membrane fraction was prepared as
224
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
described previously w17,18x. Briefly, the cells from a 100 mm diameter plastic Petri plate were scraped and centrifuged Ž700 = g . for 10 min and placed on ice. The cell pellets were resuspended in buffer containing 250 mM sucrose, 5 mM NaN3 , 2 mM EGTA, 100 mM phenylmethylsulfonylfluoride, 1 m M leupeptin, 1 m M pepstatin and 20 mM Hepes ŽpH 7.4. and homogenized using 20 strokes of a Dounce homogenizer. The homogenate was centrifuged at 760 = g for 5 min after which the supernatant was centrifuged at 31 000 = g for 60 min. The pellet Ži.e., the crude plasma membrane. was resuspended in the homogenization buffer stored at y708C and employed for western analysis. Plasma membrane markers were assayed to verify plasma membrane enrichment. Ouabain sensitive Naq-ATPase w19x activities for the homogenate were 16 " 9.2, 12 " 3.2 and 15 " 3.2 for control, insulin-treated and ASP-treated cells, respectively. For the crude plasma membranes, activities were 57 " 12, 57.6 " 19 and 54.7 " 16 for control, insulin-treated and ASP-treated cells, respectively, measured as phosphatermg protein per 60 min Ž n s 2 experiments. which ranges from 3- to 4.8-fold enrichment in the crude plasma membrane fraction. 2.6. SDS gel electrophoresis and western blots Purified human erythrocyte glucose transporter was prepared as described elsewhere w10x. A crude extract of rat brain cortex was employed as a standard for Glut 3 Ža gift from the laboratory of Dr. S. Cushman.. Western blot analysis was carried out as follows: 1. samples were solubilized in Laemmli w20x sample buffer and electrophoresis was performed on 10% polyacrylamide slab gels as described earlier w17,18x. 2. Proteins were blotted onto nitrocellulose paper Ž 0.2 m m; Schleicher and Schullx that was blocked with 1% BSA Ž 1 h at 208C.. 3. This was followed by incubation with specific antibody Žovernight at 48C.. 4. Subsequently, 125 I-labelled protein A at 1.2 = 10 6 dpmr10 ml in blocking buffer containing 0.1% Triton X-100 was added for 1 h at 208C. 5. Autoradiography was performed on Kodak XAR film ŽEastman Kodak, Rochester, NY. exposed at y858C overnight. Spots on the autoradiograms were marked using the developed X-ray film, cut out and counted in a gamma counter for quantitation.
2.7. Statistics All results are expressed as average" S.E.M. Student’s t-test Žtwo-tailed. for two means was used as a comparison between two different treatment groups. A one-way ANOVA was used when comparing more than two means.
3. Results The data in Fig. 1 demonstrate the effect of increasing insulin exposure ŽFig. 1, top. and increasing insulin concentration ŽFig. 1, bottom. on differentiated L6 myotubes. The effect of insulin on 2-DG transport was maximal after 1 h exposure at an insulin concentration of 0.067 m M ŽFig. 1, top. where basal glucose transport is 249 " 54 pmol 2-DGrmg protein per 5 min. These data are similar to those previously reported by Klip and colleagues employing the same cell type and mode of induction of differentiation w17,18x. As shown in Fig. 1 bottom, this effect of insulin was concentration dependent. Having established that this muscle cell model system was behaving as expected, we sought to determine whether these cells would respond to ASP. The data in Fig. 2 top, clearly indicate that L6 myotubes respond very well to ASP Ž 2.8 m M. exhibiting an increase in 2-DG transport after 30 min exposure which increased up to 1–2 h. In fact, the ASP response curve was nearly identical to the insulin response curve over the 2 h time frame of the experiment as also shown in Fig. 2. Of considerable interest was the observation that the combined effects of ASP Ž2.8 m M. plus insulin Ž0.067 m M. exhibited an increase in 2-DG transport well above the individual treatments suggesting that the responses were additive. This ASP response by the L6 cells was dose dependent with the greatest response to ASP occurring at an ASP concentration of 2.8 m M ŽFig. 2, bottom. Ž half maximal concentrations 0.5 m M.. As shown in Table 1, ASP had no effect on the K m of 2-DG transport in myotubes Ž1.1 mM vs. 1.8 mM in the control vs. ASP groups respectively. while the Vmax for 2-DG transport was increased by 183% Ž4.02 vs. 1.42 nmolrmg protein per 30 s; ASP vs. Control respectively. . We then determined if the response to ASP of
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
225
undifferentiated myoblasts and differentiated myotubes differed. The data in Table 2 compare the responses of myoblasts and myotubes to insulin, ASP, and a combination of the two factors. The myoblasts exhibited marginal increases in glucose transport to all factors Ža range of 12% to 15% increase above
Fig. 1. Time- and concentration-dependent effect of insulin on glucose transport in L6 myotubes. Myoblasts were grown to 75% confluence, then differentiated over 4–5 days to myotubes. Top panel: myotubes were exposed to 0.067 m M insulin for 15 min to 2 h and 2-deoxy-glucose Ž2-DG. transport was then assessed. Results are expressed as 2-DG Uptake Ratio calculated as Insulin treatedrcontrol for an average of ns 2 experiments Žtriplicate plates in all experiments. where basal glucose transports 249" 54 pmol 2-DGrmg protein per 5 min. Bottom panel: myotubes were exposed for 1 h to increasing concentrations of insulin as indicated followed by assessment of glucose transport. Results are shown for a representative experiment Žtriplicate plates for each point. and are expressed as 2-DG Uptake Ratio calculated as Insulin treatedrcontrol where basal glucose transports 586" 33 pmol 2-DGrmg protein per 5 min.
Fig. 2. Time- and concentration-dependent effect of ASP on glucose transport in L6 myotubes. Myoblasts were differentiated as in Fig. 1. Top panel: myotubes were exposed to 2.8 m M ASP Žv . or 0.067 m M insulin Ž ) . Žor both together ŽB. for 30 min to 2 h and 2-deoxy-glucose Ž2-DG. transport was then assessed and compared to Control ŽP.. Results are expressed as pmoles 2DGrmg cell protein ŽP. per 5 min for an average of ns 3 experiments Žtriplicate plates in each experiment.. Bottom panel: myotubes were exposed for 1 h to increasing concentrations of ASP followed by assessment of glucose transport. Results are expressed as pmoles 2-DGrmg cell protein ŽP. per 5 min for an average of ns 4 experiments; P - 0.05 by one-way ANOVA.
control. where basal levels of glucose transport were 2.279 nmol 2-DGrmg cell protein, 5 min for myoblasts and 4.222 for myotubes. The responses of the myotubes, however, were significantly elevated with all treatments versus the myoblasts Ža range of 44% to 77% increase above control; P - 0.01 by one-way ANOVA. . Since 2-DG transport was increased in response to ASP, we sought to determine the precise
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
226
Table 1 The effect of ASP on the K m and Vmax of 2-deoxy-glucose Ž2-DG. Treatment
2-DG transport
Ž1 h.
K m ŽmM.
y ASP q ASP Ž2.8 m M.
1.1"0.19 1.8"0.16
Vmax Žnmol 2-DGr mgP per 30 s. )
1.42"0.21 4.02"0.56
))
)
No significant difference P ) 0.05. Significant difference P - 0.05. Cells were differentiated as described in Section 2, serum deprived overnight and exposed to ASP Ž2.8 m M. for 1 h. Seven different 2-DG concentrations were employed Ž0.05 mM to 5.0 mM.. The uptake time was 30 s Žlinear for all concentrations. and saturable sugar uptake was determined by subtracting the diffusional uptake using L-glucose w3x. Data is mean"S.E.M. and represents the average of 2 experiments, with duplicate plates at all concentrations. ))
mechanismŽ s. that were involved in this ASP induced increase in glucose transport. L6 myotubes were exposed to insulin Ž 0.067 m M. or ASP Ž2.8 m M. for 1 h and a crude plasma membrane preparation was isolated using identical procedures as described elsewhere w17,18x. Plasma membrane extracts were then separated by molecular weight on SDS polyacrylamide gel electrophoresis and probed using anti-glucose transporter antibodies. The data in Fig. 3 Žpanel A1, A2, A3. show the results of the Western blot Table 2 The effect of ASP and insulin on 2-deoxy-glucose Ž2-DG. transport ratios in myoblasts and myotubes Treatment
2-DG transport ratios
Ž1 h.
myoblasts
myotubes
ASP Ž2.8 m M. ASP Ž2.8 m M.qinsulin Ž0.067 m M. Insulin Ž0.067 m M.
1.12"0.04 1.14"0.02
1.55"0.08 1.70"0.08
)
1.15"0.01
1.47"0.06
)
)
)
One-way ANOVA indicated that all transport ratios were significantly different from those observed for the myoblasts Ž F s 19.53; P - 0.01.. Myoblasts were grown to near confluence Ž75%. and myotubes were induced in 2% FBS MEM for 5 days as described in Section 2. 2-DG transport was determined after 18 h in serum-free DMEM Ž0.05 mM 2-DG for minutes.. All treatments were for 1 h. Data represent the average of 6 experiments"S.E.M. where basal glucose transport is 2.279 Žmyoblasts. and 4.222 Žmyotubes. nmolrmg protein per 5 min.
Fig. 3. Western analysis of ASP and insulin effects on GLUT 1, GLUT 3, and GLUT 4 glucose transporters in L6 myotube plasma membrane preparations. Myoblasts were differentiated as in Fig. 1. L6 myotubes were exposed to ASP Ž2.8 m M. or insulin Ž0.067 m M. for 1 h and a crude plasma membrane was then isolated. Western analysis was performed and the autoradiograms for GLUT 1 Žpanel A1., GLUT 3 Žpanel A2. and GLUT 4 Žpanel A3. are shown. The bands on the Western blots were cut out and counted in a gamma counter for quantitation Žpanels B1, B2 and B3 represent the GLUT1, GLUT3 and GLUT4 Western blots, respectively. Lanes a, b and c in all panels represent ASP, insulin and control respectively. Results are expressed as 125 I protein A Žcpm. for an average of 5 to 6 experiments each"S.E.M., ) P - 0.05 vs. control.
analysis of GLUT1, GLUT3 and GLUT4 transporters in this plasma membrane preparation. In Fig. 3 Žpanel B1. there was a clear increase in the amount of GLUT 1 transporter in the membranes in response to stimulation by insulin Žlane b, 47% average increase, P - 0.05. and, importantly, in response to ASP Žlane a, 43% average increase, P - 0.05.. A predominant GLUT 1 doublet was apparent in the ASP treated group but was not noticeable in any other treatment group. In this same series of experiments, the membrane preparations were probed for the GLUT 4 transporter ŽFig. 3, panel A3. . The data indicate that both insulin Žlane b. and ASP Žlane a. significantly increased the GLUT 4 transporter content of the membrane preparations by an average 26% and 30%, respectively Ž P - 0.05 for both ASP and insulin vs. control.. No doublets of the GLUT 4 transporter were observed in any experiment. Continued analysis was performed on the same membrane preparations for
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
the presence of the GLUT 3 transporter ŽFig. 3, panel A2.. The data in Fig. 3 Žpanel B2., show that ASP and insulin Žlanes a and b respectively. increased the GLUT 3 transporter content of the membranes by an average 49% Ž P - 0.05. and 53% Ž P - 0.05., respectively. Again, no doublets of this transporter were observed in any experiments. The stimulation of glucose transport done in parallel with the translocation experiments yielded a 67% average increase in the insulin group and a 40% average increase in the ASP group Žboth treatments were significantly different from the control group, two-tailed t-test, P - 0.01, n s 6.. Having observed short-term effects of ASP on 2-DG transport, we next investigated the effects of long-term exposure Žup 24 h. by ASP, insulin or a combination thereof on 2-DG transport. Over the 24 h period in serum-free medium basal glucose transport decreases slightly although not significantly. This is probably due to the extended serum-free incubation in a high concentration of glucose. The data seen in Table 3 clearly demonstrate that the effects of ASP on 2-DG transport are present from 2–24 h Ž 38% increase after 2 h, 96% after 8 h, 78% after 8 h and 97% after 24 h.. In addition, the effects of insulin were also elevated over these same time frames Ž45%; 93%; 106% and 125%, respectively.. Further, the effects of the combined treatments were significantly elevated over either treatment group suggesting a continued enhancement of 2-DG transport over long time frames Ž 96%; 160%; 199% and 232%; increase over basal at 2–24 h, respectively.. The total cellular content of the various glucose transporters was determined after 24 h exposure to the different treatments ŽFig. 4.. The GLUT1 trans-
227
Fig. 4. Western analysis of ASP and insulin effects on total cellular GLUT1, GLUT3 and GLUT4 glucose transporters in L6 myotubes. L6 myotubes were differentiated as described in Section 2, incubated overnight in serum-free medium and then incubated in basal medium Žopen bars., with ASP Ž2.8 m M, solid bars., insulin Ž0.067 m M, striped bars. or both Žcross-hatched bars. for 24 h. Cells were harvested and total glucose transporters assessed by Western analysis. Results are presented for an average of 3 experiments"S.E.M., where ) s P - 0.05 versus control and as P - 0.05 for ASPqinsulin versus ASP alone.
porter showed a trend towards increased content following treatment with ASP, however, the differences were not significant. ASP alone had no effect on total cellular content of GLUT3 or GLUT4 although ASP did increase translocation of GLUT3 and GLUT4 ŽFig. 3. and overall glucose transport. However, ASP did enhance significantly the insulin-stimulated increase on total cellular transporter content for both GLUT3 Ž166% ASP q insulin versus 143% insulin alone, P - 0.05 where basal s 100%. and GLUT4 Ž276% ASP q insulin versus 205% insulin alone, P - 0.05 where basal s 100%.. Thus, ASP demonstrates both short-term and
Table 3 The effect of long-term exposure to ASP and insulin in L6 myotubes2-DG transport Žpmolrmg protein per 5 min. Time Žh. 2 4 8 24 a
Control 852 " 36 804 " 39 708 " 66 607 " 84
qASP Ž2.8 m M. 1180 " 138 1578 " 440 1264 " 66 1201 " 146
a a
qInsulin Ž.067 m M. a
1238 " 80 1545 " 103 1453 " 164 1374 " 186
a a
ASP q insulin Ž2.8 m M q .067 m M. 1667 " 137 a,b 2091 " 489 a,b 2120 " 314 a,b 2021 " 371 a,b
Significant difference vs. control Ž P - 0.05.. Significant difference vs. ASP- or insulin-treated P groups Ž P - 0.05.. Cells were treated as described in Section 2 for the times indicated above. 2-DG transport was assayed at a concentration of 0.05 mM 2-DG for 5 min. Data represent the average of three experiments Žtriplicate plates in each experiment. " S.E.M. Statistical analysis was a one-way repeated measures ANOVA Ž F s 35.87 at 2 h; 23.15 at 4 h; 19.91 at 8 h and 11.5 at 24 h..
b
228
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
long-term effects on glucose transport via effects on total cellular glucose transport content and translocation of glucose transporters to the plasma membrane.
4. Discussion The present studies confirm previous data as to the effect of insulin on glucose transport in L6 myotubes w17,18,21x. Importantly the data also demonstrate that ASP markedly stimulates specific membrane transport of glucose in L6 myotubes, an effect which occurred at physiological concentrations of ASP. As with cultured human skin fibroblasts, the effect of ASP on this process was concentration and time dependent and was comparable to that of insulin. However, the effect of ASP was independent of insulin and additive to insulin. Given the evidence that suggests that different activation cascades are employed by ASP and insulin, this independence and additivity of effects should not be surprising w7x. The effect of both ASP and insulin on glucose transport was pronounced in the differentiated myotubes, but there was little effect in the undifferentiated myoblasts pointing to a selective differentiation induced response to both factors. Interestingly, we have previously shown that differentiated human adipocytes are more responsive than undifferentiated human preadipocytes with respect to ASP stimulation of triglyceride synthesis w2x. Experiments were carried out to determine the specific mechanism by which ASP caused the membrane transport of glucose to increase. Our data clearly demonstrate that ASP treatment resulted in a significant increases in the myotube plasma membrane content of GLUT1, GLUT3 and GLUT4 glucose transporters indicating that translocation of all three glucose transporters was effected by exposure to physiological levels of ASP. Insulin stimulation of the GLUT1, GLUT3 and GLUT4 glucose transporters was 47%, 64% and 26% greater than control membranes while ASP stimulation of the same transporters, respectively, was 43%, 62% and 30% greater than control membranes. The extent of translocation paralleled 2-DG uptake in parallel cultures treated with insulin Ža 67% average increase. or ASP Ža 40% average increase.. The protein kinase C pathway has been demonstrated to be involved in the signal trans-
duction mechanism by which ASP activates the appropriate intracellular metabolic responses Ž triglyceride synthesis and glucose transport. w7x and may be involved in translocation of glucose transporters. Long-term exposure of myotubes to ASP Ži.e., for up to 24 h. resulted in continued elevated rates of glucose transport ŽTable 3.. These results suggest that ASP as well as insulin can effect long-term increases in glucose transport in muscle. Our previous observations on glucose transport after long-term exposure of human fibroblasts to ASP w8x contrast with the present effects of ASP on muscle cell cultures. In those studies, ASP stimulation of glucose transport peaked 2 h post ASP exposure, was decreasing by 8 h and, after 24 h exposure to ASP, glucose transport rates were no longer increased. In the present study, glucose transport is elevated up to 24 h. Of considerable interest is that the ASP q insulin group is significantly elevated at 8 and 24 h post treatment when compared to ASP or insulin alone. The data in Fig. 4 which reflects the total amount of cellular glucose transporters present after 24 h treatment indicate that the elevations in transport observed are associated with increased cellular levels of GLUT3 and GLUT4. Thus, the differences observed most likely reflect the differences between the effect of ASP on the undifferentiated fibroblast vs. the differentiated state of L6 myotubes. That the biologic effect persists over prolonged periods with the differentiated cell further points to physiologic relevance of the observations. Our previous work has focused on the role of ASP in human triglyceride metabolism. Thus, following an oral fat load, but not after an oral glucose load, plasma ASP levels gradually rise and fall, roughly paralleling the changes in plasma triglycerides w22x. In normals, the rate at which triglycerides were cleared was related to both the fasting and peak plasma levels of ASP consistent with the hypothesis that plasma ASP is a determinant of peripheral triglyceride clearance. ASP stimulates triglyceride synthesis markedly in differentiated human adipocytes and preliminary results indicate that this is also true of glucose transport in differentiated human adipocytes w23x. Adipocytes from patients with hyperapoB synthesize triglycerides less rapidly than normal w24x and fibroblasts from such patients respond
Y. Tao et al.r Biochimica et Biophysica Acta 1344 (1997) 221–229
less well to ASP w6x, suggesting that dysfunction of the adipsin-ASP pathway may be important in the pathogenesis of this dyslipoproteinemia. By contrast, in patients with gynoid obesity, adipocytes remain responsive to ASP w25x and plasma levels of ASP are markedly elevated in such patients w26,27x. Studies in cultured cells indicate that ASP acts via a second messenger system involving protein kinase C w7x. This cell triggering is likely mediated through specific interaction with a cell surface receptor w6x. Studies to identify and characterize the ASP receptor are ongoing. Given the similarities which have been demonstrated between the mechanisms responsible for carrier-mediated glucose transport in human muscle cells in culture and L6 myotubes w28x, the present data raise the possibility that ASP may also act in vivo on muscle, a traditional insulin target tissue, as well as adipocytes. If this is the case, then ASP may play a broader role in glucose transport than previously recognized.
Acknowledgements This research was supported by a grant from the Medical Research Council of Canada to Dr. K. Cianflone. Dr. K. Cianflone is the recipient of a scholarship of the Heart and Stroke Foundation of Canada and ‘les Fonds de la Recherche en Sante´ du Quebec’. ´
References w1x Baldo, A., Sniderman, A.D., St-Luce, S., Avramoglu, R.K., Maslowska, M., Hoang, B., Monge, J.C., Bell, A., Mulay, S. and Cianflone, K. Ž1993. J. Clin. Invest. 92Ž3., 1543– 1557. w2x Cianflone, K., Roncari, D.A.K., Maslowska, M., Baldo, A., Forden, J. and Sniderman, A.D. Ž1994. Biochemistry 33, 9489–9495. w3x Cianflone, K., and Maslowska, M. Ž1995. Eur. J. Clin. Invest. 25, 817–825. w4x Choy, L.N., Rosen, B.S. and Spiegelman, B.M. Ž1992. J. Biol. Chem. 367, 12736–12741.
229
w5x Sniderman, A.D. and Cianflone, K. Ž1995. in Yearbook of Endocrinology, pp. xix–xxxvii, Mosby, Chicago, IL. w6x Cianflone, K., Maslowska, M. and Sniderman, A.D. Ž1990. J. Clin. Invest. 85Ž3., 722–730. w7x Baldo, A., Sniderman, A.D., St-Luce, S., Zhang, X.J. and Cianflone, K. Ž1995. J. Lipid Res. 36, 1415–1426. w8x Germinario, R., Sniderman, A.D., Manuel, S., Pratt, S., Baldo, A. and Cianflone, K. Ž1993. Metabolism 40Ž5., 574–580. w9x Yasruel, Z., Cianflone, K., Sniderman, A.D., Rosenbloom, M., Walsh, M. and Rodriguez, M.A. Ž1991. Lipids 26Ž7., 495–499. w10x Baldwin, S.A., Baldwin, J.M. and Lienhard, G.E. Ž1982. Biochemistry 21, 3836–3842. w11x Pratt, S.E., Colby-Germinario, S.P., Manuel, S. and Germinario, R.J. Ž1994. J. Cell Physiol. 161, 580–588. w12x Germinario, R.J., Rockman, H., Olveira, M. and Manuel, S. Ž1982. J. Cell Physiol. 112, 367–372. w13x Germinario, R.J., Lakshmi, T.M. and Thirion, J.P. Ž1989. J. Cell Physiol. 138, 300–304. w14x Germinario, R.J., Kristof, A., Chang, Z. and Manuel, S. Ž1990. J. Cell Physiol. 145, 318–323. w15x Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. Ž1951. J. Biol. Chem. 193, 265–275. w16x Hugli, T.E. Ž1975. J. Biol. Chem. 250, 8293–8301. w17x Mitsumoto, Y., Burdett, E., Grant, A. and Klip, A. Ž1991. Biochem Biophys. Res. Commun. 175, 652–659. w18x Bilan, P., Mitsumoto, Y., Maher, F., Simpson, I. and Klip, A. Ž1992. Biochem. Biophys. Res. Commun. 186, 1129– 1137. w19x Hildago, C., Gonzales, M.E., and Lagos, R. Ž1983. Arch. Biochem. Biophys. 253, 279–286. w20x Laemmli, U.K. Ž1970. Nature 227, 680–685. w21x Mitsumoto, Y. and Klip, A. Ž1992. J. Biol. Chem. 267, 4957–4962. w22x Cianflone, K., Vu, H., Walsh, M., Baldo, A. and Sniderman, A.D. Ž1989. J. Lipid Res. 30Ž11., 1727–1733. w23x Maslowska, M., Sniderman, A.D., Germinario, R. and Cianflone, K. Ž1995. Clin. Invest. Med. 18ŽSuppl. B., B42. w24x Teng, B., Forse, A., Rodriguez, M.A. and Sniderman, A.D. Ž1988. Can. J. Pharmacol. Physiol. 66Ž3., 239–242. w25x Walsh, M.J., Sniderman, A.D., Cianflone, K., Vu, H., Rodriguez, M.A. and Forse, R.A. Ž1989. J. Surg. Res. 46, 470–473. w26x Sniderman, A.D., Cianflone, K. and Eckel, R.H. Ž1991. Int. J. Obes. 15Ž5., 333–336. w27x Cianflone, K., Kalant, D., Marliss, E.B., Gougeon, R. and Sniderman, A.D. Ž1995. Int. J. Obes. 19, 604–609. w28x Sarabia, V., Ramlal, T. and Klip, A. Ž1990. Biochem. Cell Biol. 68, 536–542.