(Received for publication, July 19, 1996, and in revised form, December 3, 1996)
From the Dipartimento di Biologia e Patologia
Cellulare e Molecolare "L.Califano" and Centro di Endocrinologia ed
Oncologia Sperimentale del C.N.R. and the ¶ Dipartimento di
Medicina Clinica e Sperimentale, "Federico II" University of
Naples Medical School, 80131 Naples, Italy
L6 myotubes expressing the constitutively active
Arg1152Gln insulin receptor (L61152)
featured a 31% increased glucose consumption as compared with L6 cells
expressing wild-type receptors (L6WT). However, insulin
treatment decreased glucose consumption of the mutant cells by 20%
while increasing that of the L6WT by 30%. In the
L6WT, insulin elicited a significant increase in glucose
transport and GLUT1 and GLUT4 plasma membrane expression, while in the
L61152, all of these functions were constitutively
activated and not further stimulated by insulin. Similarly, glycogen
content and glycogen synthase activity were increased by 80 and 125%,
respectively, in the L61152 versus the
L6WT and unaffected by insulin (while a 2-fold increase was
measured in insulin-exposed L6WT). Glucose oxidation and
pyruvate dehydrogenase activity were also 25% higher in the mutant
compared with the L6WT. However, in the L61152,
both functions decreased by 35% in response to insulin (while increasing by 60 and 80%, respectively, in the L6WT).
Similarly as in the L61152, in vivo, forearm
glucose uptake in IR1152 patients was 2-fold higher than in
control subjects. This difference was not accounted for by higher
plasma glucose levels. We conclude that, in skeletal muscle, glucose
storage and oxidation are differentially impaired by the expression of
IR1152, suggesting that their regulation by insulin
involves divergent signaling pathways. Muscle expression of
IR1152 may contribute to impairing glucose tolerance in
IR1152 individuals.
Glucose tolerance is largely determined by insulin stimulation of glucose utilization by the liver and skeletal muscle and by insulin suppression of liver glucose output (1, 2). Decreased glucose utilization and increased glucose production by these tissues play an important causal role in generating hyperglycemia in non-insulin-dependent diabetes mellitus (NIDDM)1 (1, 3). However, the precise molecular mechanisms leading to these defects, as well as the relative contribution of muscle and liver to hyperglycemia in NIDDM are still unclear (1, 3).
A number of insulin receptor mutations have been described in NIDDM individuals (4, 5). Most of these defects have been studied either in cultured lymphocytes/skin fibroblasts from the patients or in transfected fibroblast-like cells. Therefore, the consequences of the mutant receptor expression in tissues relevant to in vivo glucose tolerance have received little attention until very recently. This is an important problem because (i) the expression of mutant insulin receptor alleles may occur in as much as 10% of the NIDDM individuals, and thus they might contribute to impaired glucose tolerance in a significant fraction of these patients (4, 5); and (ii) selective expression of defective receptors in liver or skeletal muscle may provide convenient models for investigating the relative contribution of these tissues to hyperglycemia in NIDDM.
In a family of individuals with NIDDM but no clinical features of the
known genetic syndromes of severe insulin resistance, we have
identified an insulin receptor point mutation leading to
Arg1152Gln substitution in the regulatory domain of
receptor
-subunits (IR1152) (6). IR1152
patients exhibit marked skeletal muscle insulin resistance in glucose
disposal as compared with most other NIDDM individuals but little
change in insulin suppression of liver glucose production (6, 7). This
suggested an important involvement of muscle in altering glucose
tolerance in these patients. The Arg1152
Gln substitution
leads to constitutive activation of receptor kinase and signaling
(8-10). In fibroblasts from the IR1152 patients and in
NIH-3T3 cells expressing IR1152, glucose and aminoacid
transport across the plasma membrane and glycogen synthase activity
exhibit high basal values and do not further increase in response to
insulin (8-11). However, as in the case of most other naturally
occurring insulin receptor mutants, the molecular mechanisms leading to
IR1152 impairment of glucose disposal by skeletal muscle
are unclear.
In the present work, this issue has been addressed by in
vitro and in vivo studies. It has been shown that, at
variance with all other mutations in the kinase domain of the insulin
receptor, the Arg1152Gln substitution results in a gain
of function enabling the receptor to signal inhibition of pyruvate
dehydrogenase activity in response to insulin. Evidence was also
obtained that insulin regulates glucose storage and oxidation in
skeletal muscle cells through divergent signaling pathways.
Preparation of plasmid DNA, agarose gel electrophoresis, restriction enzyme digestion, and DNA sequencing were performed by standard methods (12). Enzymes were from Boehringer Mannheim (Kvistegard, Denmark) or Pharmacia Biotech Inc. (Hillerod, Denmark). All radiochemicals as well as monoclonal Ig2 phosphotyrosine antibodies were from Amersham Corp. (Milano, Italy). mAb3 IR antibody was purchased from Oncogene Science (Manhasset, NY). The transfection reagent (DOTAP) was purchased from Boehringer Mannheim. Media and serum for tissue culture were from Life Technologies, Inc. (Grand Island, NY), electrophoresis and Western blot reagents were from Bio-Rad (Richmond, CA), and sulfo-N-hydroxysuccinimide long chain biotin was from Pierce. All other chemicals were from Sigma.
Transfection, Cell Culture and Mutant Characterization, Insulin Binding, Insulin Receptor Metabolic Labeling, and PhosphorylationWild-type and mutant insulin receptor constructs have been previously described (8). The L6 skeletal muscle cells were seeded (5-9 × 103 cells/cm2) and grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 2% fetal bovine serum as described previously (14). Under these cultural conditions, the myoblasts reach the confluence and then spontaneously differentiate into myotubes. Cationic liposome-mediated transfection of L6 myoblasts was performed according to Levine et al. (15) and to the DOTAP reagent manufacturer instructions. G-418 (Life Technologies, Inc.) was used at the effective dose of 0.8 mg/ml. Individual G-418-resistant clones were isolated and screened by 125I-insulin binding. Neither the cell clones expressing the wild-type or those expressing the mutant receptors exhibited significant change in morphology compared with the untransfected cells either at the myoblast or the myotube stage of differentiation. Based on fusion index (the relative proportion of nuclei in myotubes and in mononucleated cells (16)) and quantitation of creatine kinase activity, the ability of the L6 myoblasts to differentiate into myotubes as well as the differentiation time course were also unchanged upon transfection (Table I).
|
Unless differently specified, all of the biochemical studies shown in this work were performed with L6 cell clones (parental and transfected) at the myotube stage. In order to minimize the likelihood of nonspecific clonal variation however, control studies were also accomplished with the pools of transfected cells. Under no circumstance were discrepancies between the pool and the corresponding cell clones observed. 125I-insulin binding was studied as described previously (17) and analyzed according to Scatchard (18). Insulin receptor metabolic labeling, autophosphorylation, and kinase activity were studied as described earlier (8). CK activity was determined in the presence of diadenosine pentaphosphate using a previously described method (19).
Glucose Disposal, 2-Deoxy-D-glucose and 3-0-methylglucose UptakeFor glucose utilization studies, L6 myotubes were maintained in DMEM supplemented with 2% serum and 100 mg/dl glucose for 6 days. The cells were then incubated at 37 °C in a humidified atmosphere containing 5% CO2 and 95% air in fresh medium in the absence or presence of 100 nM insulin for 12 h. Glucose disappearing from the culture medium (determined by measurement of glucose concentrations (20)) in that period was considered to be utilized by the cells. 2-Deoxy-D-glucose (2-DG) uptake was determined as described in Ref. 21. 3-O-methylglucose uptake was analyzed as described earlier (22).
Detection of Glucose Transporters at the Cell Surface and in Total Cell ExtractsCell surface expression of glucose transporters was analyzed according to Levy-Toledano, et al. (23). Briefly, myotubes were pre-incubated in serum-free DMEM supplemented with 0.25% BSA for 24 h and then further incubated in the absence or presence of 100 nM insulin for 1 h. The cells were rinsed and treated with 0.5 mg/ml N-hydroxysuccinimide long chain biotin in PBS containing 0.1 mM CaCl2, 1 mM MgCl2, pH 7.4 (biotinylation buffer), for 30 min at 4 °C. The biotinylation reaction was quenched by adding 15 mM glycine in biotinylation buffer. The cells were then lysed in a solution containing 1% Triton X-100, 50 mM Hepes, pH 7.4, 10 mM Na4P2O7, 100 mM NaF, 4 mM EDTA, 2 mM Na3VO4, 2 mM phenylmethylsulfonyl fluoride, and 0.2 mg/ml aprotinin, and proteins were precipitated with either GLUT1 or GLUT4 antibodies. Precipitated proteins were separated by 10% SDS-PAGE and transferred on nitrocellulose filters by blotting at 180 mA for 1 h (transfer buffer contained 25 mM Tris, 192 mM glycine, 20% methanol). Biotinylated transporters were finally revealed by incubation with peroxidated streptavidin and detection of chemioluminescence by autoradiography.
For determination of total transporter content, cell lysates (prepared as described above) were subjected to 10% SDS-PAGE and Western blotting as described above. Filters were incubated with GLUT1 or GLUT4 antibodies for 14 h at 4 °C and then with peroxidated anti-antibodies for 1 h at room temperature. Transporters were finally revealed by detection of chemioluminescence by autoradiography.
Cell Glycogen Content and Glycogen Synthase ActivityGlycogen content was determined as described by Keppler et al. (24). Briefly, myotubes were maintained in serum-free DMEM supplemented with 0.25% BSA in the absence or the presence of 100 nM insulin for 12 h. Cells were then collected in 0.6 N HClO4, homogenized using a glass-teflon potter, and centrifuged at 1500 rpm for 10 min at 4 °C. Aliquots of the homogenate were incubated with 0.033 M KHCO3 and 9 mg/ml amyloglucosidase in 0.2 M acetate buffer, pH 4.8, for 2 h at 40 °C. The reaction was stopped by addition of 0.6 N HClO4 and centrifugation at 15,000 rpm at 4 °C for 15 min. Glucose concentration was determined with a Beckmann glucose analyzer.
Glycogen synthase activity was determined by a modification of the
method of Thomas et al. (25). For this assay, myotubes were
preincubated in Hepes buffer, pH 7.4, for 3 h at 37 °C. The cells were then exposed to 100 nM insulin for 30 min at
37 °C, rinsed with ice-cold 100 mM NaF, 10 mM EDTA, collected, and centrifuged at 2,000 rpm at 4 °C
for 10 min. The pellet was resuspended in 100 mM NaF and 10 mM EDTA, sonicated for 10 s at 300 watts at 4 °C,
and further centrifuged at 2,000 rpm for 10 min at 4 °C. 20-µl
aliquots of the supernatants were added to 60 µl of a reaction mixture containing 40 mM Tris-HCl, pH 7.8, 25 mM NaF, 20 mM EDTA, 10 mg/ml glycogen, 7.2 mM uridine 5-diphosphate-glucose (UDPG), and 0.05 µCi
[14C]-UDPG in the absence or the presence of 6.7 mM glucose 6-phosphate (final concentrations). The
incubation was prolonged for 20 min at 30 °C and the reaction
terminated by precipitating 75 µl of the mixture on 2 × 2 cm
squares of filter paper with cold ethanol. The filter papers were
extensively washed with cold ethanol, and radioactivity was counted in
a Beckman scintillation counter. Enzyme activity was expressed as
percent of the glucose 6-phosphate-independent form.
Glucose oxidation was determined by a modification of
the method of Lowe and coworkers (26). Briefly, for these experiments, cells were grown in 50-ml flasks and preincubated in serum-free DMEM
for 18 h before the experiment. This medium was substituted with
Joklik medium supplemented with 25 mM NaHCO3,
5.55 mM [U-14C]glucose, 1.2 mM
MgSO4, 0.5 mM CaCl2, 10 mM Hepes, pH 7.4, in the absence or the presence of 100 nM insulin, and the incubation was prolonged for 20 min at
30 °C after capping the flasks with rubber stoppers containing a
hanging well filled with rolled filter paper. 0.4 ml 1 M
hyamine hydroxide in methanol was then injected through the rubber
stoppers into the hanging wells followed by injection into the
incubation medium of 0.4 ml 10% HClO4. The flasks were
allowed to sit for 2 more h at 37 °C. The filter papers were then
removed and counted in a -counter. Pyruvate dehydrogenase activity
was assayed as the release of 14CO2 from
[1-14C]pyruvic acid according to Seals and Jarett (27).
For these assays, 100-mm cell dishes were incubated for 10 min at
37 °C in the absence or the presence of 100 nM insulin
in DMEM with 10 mM Hepes, pH 7.4, 0.2% BSA. This medium
was supplemented with either 0.5 mM CaCl2, 10 mM MgCl2 (for determination of total pyruvate dehydrogenase (PDH) activity), or 10 mM dichloroacetate and
10 mM NaF (for determination of basal activity (28)). Cell
extracts were obtained according to Benelli et al. (29), and
50-µl aliquots were added to 200 µl of 50 mM Tris-HCl,
pH 7.4, 50 µM CaCl2, 50 µM
MgCl2 in 17 × 100 mm polypropylene tubes and
incubated. The assay was then initiated by addition of 1 mM
dithiothreitol, 0.1 mM coenzyme A, 0.25 mM
pyruvic acid labeled by 0.125 µCi [1-14C]pyruvic acid
(specific activity 9.8 mCi/mM), 0.5 mM
-NAD,
0.1 mM L-cocarboxylase (final concentrations).
The tubes were immediately capped with a rubber stopper through which
was suspended a plastic well containing a small roll of filter paper.
After 5 min at 37 °C, the reactions were stopped by injecting 0.4 ml
of 3M H2SO4 through the rubber
stopper into the reaction mixture. 14CO2 was
collected for 1 h by injecting 0.2 ml of 1 M hyamine
hydroxide through the rubber stopper onto the filter paper in the
center wells. Radioactivity in the filters was determined in a
scintillation counter. Blank values were determined by using boiled
extracts under the same assay conditions as above and subtracted from
all other values. Results were expressed in
[14C]O2 nM/min/mg extract
protein. Lactate concentration in the culture medium was determined as
by Hohorst (30).
For these assays, 100-mm dishes of myotubes were deprived from serum for 18 h and stimulated with 100 nM insulin at 37 °C for 30 min. The cells were then scraped in 5 ml of 6M HClO4, homogenized in a glass-teflon homogenizer, and centrifuged at 3,000 g for 10 min at 4 °C. Supernatants were saved, and the pellets were resuspended in 1 ml 6M HClO4 and further centrifuged as above. The supernatants were pooled, brought to pH 3.5 with 5 M K2CO3, and chilled on ice. Glucose 6-phosphate concentrations were determined by A339 nm on aliquots (1 ml) of the deproteinated supernatants upon addition of 100 µl of 0.4 M triethanolamine, 20 mM NADP, 0.5 M MgCl2, and 0.25 mg/ml glucose 6-phosphate dehydrogenase and incubation for 5 min at 37 °C (31). Blank values were obtained by incubating the samples in the absence of glucose-6-phosphatase and subtracted from all of the other values.
In Vivo StudiesThe subjects were studied in the postabsorptive state after a 15-17-h overnight fast. Glucose utilization in the forearm was calculated by subtracting the glucose concentration in a deep antecubital vein of the forearm (cannulated retrogradedly) for glucose concentration measured in the brachial artery and multiplying this difference by the forearm flow measured by the dye technique (32).
The molecular mechanisms leading to the abnormal glucose disposal in muscle of the IR1152 individuals (6) were addressed by transfecting the mutant IR cDNA in the L6 cultured skeletal muscle cells. These cells maintain the ability to differentiate in culture and have been widely used as a cell model for studies on insulin action. Clonal cell lines were screened for expression of 125I-insulin binding, and several cell clones were isolated expressing either the human wild-type or the mutant IRs. Insulin sensitivity of these cells increased linearly with an IR number up to 4 × 104 receptors/cell, whereas linearity disappeared with higher receptor expression. Therefore, six clones expressing 3.2, 1.8, and 0.9 × 104 wild-type IRs/cell (WT1, WT2, WT3) and 3.5, 1.9, and 1.0 × 104 mutant IRs/cell (Mut1, Mut2, Mut3) were studied in detail (Table II). Based on Scatchard analysis (18), all of these clones exhibited dissociation constants (Kd) for insulin between 1.2 and 2 nM. This is similar to the Kd values of the WT and mutant receptors expressed in other cells and to that of the endogenous IR measured in untransfected L6 cells. Thus, all transfected cell clones analyzed in this work exhibited normal insulin binding affinities in addition to comparable receptor levels.
|
To ensure that IR mutants were properly processed and transported to
the cell surface, extracts were prepared from cells metabolically labeled with [35S]methionine. The radiolabeled IRs were
then immunoprecipitated with anti-IR mAb3. In the two cell lines
expressing the mutant IRs as well as in those expressing the WT IRs
(Fig. 1, lanes B-G), these antibodies
immunoprecipitated two proteins migrating at Mr = 130,000 and 95,000, which correspond to IR and
-subunits, respectively. Based on laser densitometry, the intensity of these bands
correlated well with the number of cell-surface receptors as measured
by insulin binding, indicating that transfected receptors were normally
synthesized and transported to the plasma membrane in the L6 cells.
and
-subunits were barely visible in untransfected L6 cells, which
express a very low number of endogenous IRs (Fig. 1, lane
A). Based on pulse-chase experiments with
[35S]methionine, there was no significant difference in
the rate of IR biosynthesis between any of the transfected and the
endogenous receptors (data not shown).
Upon in vivo autophosphorylation, receptor -subunits
migrated as 95 kDa bands on SDS gels (Fig. 2,).
Phosphorylation increased by 2- to 5-fold upon insulin stimulation of
untransfected and wild-type cells (Fig. 2, lanes A-F). In
contrast, the mutant receptor exhibited almost no insulin-stimulated
phosphorylation when expressed in L6 cells (Fig. 2 lanes G
and K). In the absence of insulin, phosphorylation of the
endogenous substrate IRS-1 was also very low in parental cells and in
those expressing WT receptors while increasing by 2- to 5-fold,
respectively, after insulin exposure. At variance, in the mutant cells,
IRS-1 exhibited high basal phosphorylation and no increase after
insulin exposure. Thus, as previously shown in human skin fibroblasts
(11) and in other transfected cell types (8-10), in the L6 cells,
IR1152 exhibited reduced autophosphorylation but
constitutively increased kinase activity toward endogenous
substrates.
To investigate the consequence of IR1152 expression on
glucose disposal by the L6 cells, extraction of glucose from the
culture medium was first quantified. Upon 12-h incubation in 100 mg/dl DMEM, untransfected cells extracted 0.85 ± 0.04 mg (value ± S.D.) of glucose/106 cells (Fig. 3,
top, bar A). Basal glucose extraction by these cells did not significantly differ from that of the WT (Fig. 3, top, bars C, E, and G). During the
same time period, however, the mutant cells had 1.02 ± 0.05 mg of
glucose disposed (31% more than did the WT cells; difference was
significant at the p < 0.01 level) (Fig. 3,
top, bars I, M, and O). A 25%
increase in basal glucose disposal was also measured with the mutant
relative to the WT cell pools (data not shown). The presence of insulin
into the culture medium increased glucose consumption by 20 and 30% in
parental and wild-type cells, respectively (change over basal was
significant at the p < 0.001; bars B, D, F,
and H). Unexpectedly, however, glucose disposal by the L6
cells expressing the mutant receptors decreased by 20% in the presence
of insulin (Fig. 3, top, bars L, N, and
P; p < 0.001). A similar result was also
observed with the mutant cell pool (not shown). It appeared, therefore, that the expression of IR1152 increased
non-insulin-dependent glucose disposal by the L6 muscle cells but reverted normal insulin regulation.
In the basal state, the initial rate of 2-DG uptake was also very
similar in untransfected and WT cells (1.2-1.3 nmol mg1
min
1; Fig. 3, middle panel, bars A, C,
E, and G). In comparison, the mutant cells exhibited a
54% increased uptake (bars I, M, and O;
difference significant at the p < 0.001 level). The
difference between WT and Mut cells was accounted for by an increase in
the Vmax of the transport in Mut cells, with no
detectable change in Km values (data not shown).
Insulin induced a 61% increase in 2-DG uptake in the parental and
wild-type cells (Fig. 3, bottom, lanes B, D, F,
and H), while in the mutant, insulin-stimulated uptake
exhibited only a small increase compared with the basal values (Fig. 3,
middle panel, bars L, N, and P). Similar to 2-DG, the basal uptake of the non-phosphorylatable glucose analog
3-O-methylglucose was 2-fold increased in mutant relative to
the WT cells and did not further increase in response to insulin (while
increasing by 2.3-fold in insulin-stimulated WT cells; Fig. 3,
bottom panel). It appeared, therefore, that glucose
transport could contribute to the augmented
non-insulin-dependent glucose disposal but not to the
decrease measured upon exposure of the mutant cells to insulin.
This issue was further investigated by first analyzing total cellular
content of GLUT4 and GLUT1 since these are important transporter
isoforms well represented in the L6 cells. The cells were solubilized,
and transporter content was studied by Western blot using specific
GLUT1 and GLUT4 mAb. Based on laser densitometry, there was little
difference in total transporter content between parental and WT cells
(Fig. 4, lanes A-C)). However, in four
independent experiments with all of the clones reported in the present
paper, mutant cells exhibited a 110 ± 5 and 61 ± 4%
(values ± S.D.) decrease in GLUT1 and GLUT4, respectively,
compared with the WT (Fig. 4, lanes D, and E;
p < 0.01), suggesting that the constitutively active
receptor down-regulated the transporters. Plasma membrane content of
the two transporters was also markedly different in wild-type and
mutant cells both in the absence and in the presence of insulin. Based
on precipitation of biotinylated cell surface proteins with GLUT1 or
GLUT4 mAb followed by streptavidin blotting and HRP detection, in five
independent experiments, GLUT4 was almost absent from the plasma
membrane of basal parental or WT cells while GLUT1 was consistently
revealed (Fig. 5 shows a representative experiment). In
comparison, Mut cells exhibited 7 ± 0.8-fold increased membrane
GLUT4 levels and a 50 ± 4% increase in that of GLUT1 compared
with the WT/parental cells (difference between Mut and Wt cells was
significant at the p < 0.001 level). GLUT4 and GLUT1 membrane content increased by 7.6 ± 1 and 0.6 ± 0.08-fold,
respectively, in both parental and WT cells after exposure to insulin
(107 M, maximally effective concentration;
difference relative to the basal was significant at the
p < 0.001 level). No further membrane recruitment
occurred upon insulin stimulation of the Mut cells either in the case
of GLUT4 or of GLUT1 so that their relative surface amounts in WT
versus the Mut cells were 1.08 and 1.20, respectively.
To further address the molecular basis of the abnormal glucose
utilization by the IR1152 cells, we have individually
analyzed the major intracellular routes of glucose disposal. As shown
in Fig. 6, top, basal glycogen contents in
parental/WT cells and in the mutants were 1.2 and 2.2 µg/mg of
protein, respectively (83% difference, significant at the
p < 0.001 level). Insulin exposure of the cells for
12 h increased glycogen content in WT cells by 2-fold but did not result in any significant further glycogen accumulation by the Mut
cells. Under basal conditions, [14C]glucose conversion
into 14CO2 was also increased by 20% in the
mutant as compared with the WT cells (Fig. 6, middle panel;
p < 0.001). This increase in basal glucose oxidation
was accompanied by a slight decrease in the ratio between the
appearance of lactate into the culture medium and glucose disposal by
the Mut cells (10%; p < 0.01; Fig. 6, bottom
panel). Interestingly, insulin exposure increased
14CO2 production of the parental and WT cells
by 35-65% (significant at the p < 0.001 level) but
reduced that of the mutant cells by about 30% (p < 0.001) (25% decrease with the Mut cell pools; data not shown). This
decrease was not due to receptor down-regulation of the mutant receptor
since the mutant does not undergo insulin-induced internalization and
down-regulation either in fibroblasts (9) or in the L6 cells (data not
shown). While the ratio between the appearance of lactate into the
culture medium and glucose disposal slightly decreased in parental and
WT cells upon stimulation with insulin (10% decrease, significant at
the p < 0.01 level), it increased by >20% in the Mut
cells (p < 0.001).
The activity of glycogen synthase and PDH were also
characterized. In parental and WT cells, glycogen synthase activity
increased by 2.1- to 2.6-fold, respectively, in response to insulin
(Fig. 7, top; p < 0.001). At
variance, in Mut cell extracts, basal activity of the enzyme was
comparable to that measured in maximally insulin-stimulated WT cells
but did not exhibit further significant increase following insulin
stimulation. Basal PDH activity in the Mut cells also appeared similar
to that measured in insulin-stimulated parental/WT cells (Fig. 7,
bottom). In parental/WT cells, insulin addition for 10 min
elicited a 60-80% increase in the activity of PDH (significant at the
p < 0.001 level). This acute effect was decreased to
30% above the basal activity when preceded by a 48-h preincubation of
the cells with 100 nM insulin (significant at the
p < 0.001 level; data not shown). At variance from the
WT/parental myotubes, in the Mut cells, acute insulin exposure resulted
in a 34% decrease in the activity of this enzyme (significant at the
p < 0.01 level).
If the inhibition of PDH and glucose to CO2 conversion were
responsible for the decreased glucose disposal occurring in the Mut
cells after exposure to insulin, one would also predict glucose intracellular accumulation since the glucose transport system appears
to be constitutively activated in these cells and glycogen accumulation
exhibits no change. To verify this hypothesis, Glu-6-P intracellular
concentrations were directly measured. Note, in Fig. 8,
that basal Glu-6-P concentrations were 0.033 nmol/l in parental/WT
cells and 0.023 nmol/l in the mutants (difference significant at the
p < 0.01 level) consistent with an accelerated glucose
consumption rate in the latter cell type. However, after exposure to
insulin, Glu-6-P concentrations increased to 0.037 nmol/l in the Mut
cells (p < 0.001) but decreased to 0.022-0.025 in the
parental and WT cells.
To further address the role of IR1152 in altering glucose clearance by skeletal muscle, we have analyzed in vivo the forearm glucose uptake in the basal state in an IR1152 individual. In two independent studies, basal uptake was 1.85 ± 0.05 (value ± range; Table III) in this individual. This value was slightly higher than that measured in the same IR1152 individual during euglycemic hyperinsulinemic clamp (1.4 mg/l/min (6)). However, similarly as in the Mut cells, the basal glucose uptake in the IR1152 individual was ~120% higher than that in control subjects whether NIDDM or non-diabetic. Hyperglycemia could not account for this difference since the NIDDM control group was significantly more hyperglycemic than the IR1152 patient. It appeared therefore that, in vivo as well as in vitro, expression of IR1152 in skeletal muscle results in increased non-insulin-dependent glucose disposal and impaired glucose utilization in response to insulin.
|
As in the case of most other naturally occurring insulin receptor mutations, the role of IR1152 in impairing glucose tolerance at the level of each individual major insulin target tissue has not been investigated. In the present work, this issue has been addressed by studying the effect of IR1152 expression on glucose disposal by skeletal muscle in vitro and in vivo. The L6 skeletal muscle cells have been used as an in vitro model since, once differentiated into myotubes, these cells acquire several characteristics of adult skeletal muscle (17, 34). In addition, these cells have been widely used for studies on insulin action (14, 17, 35-38), and therefore they represent a well characterized model system for these aims.
Similarly as in human fibroblasts and transfected NIH-3T3 cells (14, 10, 11), once expressed in the L6 myotubes, IR1152 exhibited no insulin-stimulated autophosphorylation but constitutively increased kinase activity toward the IRS-1 endogenous substrate. IR1152 expression also decreased total GLUT1 and GLUT4 glucose transporters into the cells suggesting down-regulation of the two transporter systems by the constitutively active IR1152 kinase. However, plasma membrane content of GLUT1 and GLUT4 was significantly increased in basal IR1152 cells (Mut) as compared with myotubes expressing WT receptors and did not further increase upon exposure to insulin. Cell-surface translocation of the transporters is an important mechanism resulting in glucose transport activation by insulin (39). It appeared therefore that, although down-regulating GLUT1 and GLUT4, IR1152 constitutively activated the two transporter systems in the L6 myotubes impairing the regulatory action of insulin. Accordingly, the initial rates of 2-DG and 3-O-methylglucose uptake in Mut myotubes were increased to levels comparable with those of maximally insulin-stimulated WT cells. It needs to be underlined however, that glucose transport activity appeared to be regulated by factors other than transporter recruitment since the increased content of transporter proteins in the plasma membrane largely exceeded the increase in glucose transport (both in the insulin-stimulated WT cells and in the basal Mut relative to the WT myotubes). As previously proposed in works with these same cells (40, 41) as well as other tissues (42), these factors may include changes in the specific activity and in glucose affinity of transporters.
As measured by the disappearance from the culture medium, glucose consumption of the Mut myotubes was also significantly higher than that of the WT cells. At variance with 2-DG uptake and transporter membrane content, however, exposure to insulin inhibited glucose consumption of the Mut cells. Hence, glucose disposal of insulin-stimulated Mut cells was reduced below that of basal WT cells. It is concluded that, (i) although unable to transduce most metabolic effects of insulin, IR1152 is capable of transducing metabolic signals leading to depression of glucose metabolism in response to insulin; and (ii) IR1152-induced activation of the glucose transport system of the cells could account for the basal increase in glucose consumption but not for the inhibition occurring upon insulin stimulation. The intracellular route involved in insulin inhibition of glucose consumption in the mutant cells did not appear to be glycogen synthesis since glycogen synthase activity of the Mut cells also appeared maximally activated, and its function did not either increase or decrease in the presence of insulin. In addition, the activity of the synthase measured in vitro appeared to reflect the activity in the intact cells since glycogen content of the mutants also was constitutively increased and unmodified by insulin. At variance from glycogen synthesis, [14C]glucose to 14CO2 conversion in the Mut cells was significantly depressed in response to insulin (while stimulated in WT and untransfected cells). It appears unlikely that the decrease in 14CO2 production by the mutant cells was mainly contributed by an increase in glycogen turnover since a similarly sized inhibition was also demonstrated in the activity of PDH in the Mut (but not in WT or parental). PDH is a major enzyme limiting glucose oxidation in muscle (43). We suggest therefore that (i) an insulin-induced decrease in 14CO2 production by the Mut cells truly reflects decreased glucose oxidation, and that (ii) insulin-induced inhibition of PDH through the IR1152 may be responsible for the observed impairment in glucose oxidation. Decreased glucose oxidation, in turn, could account for the decrease in glucose consumption occurring after insulin exposure of the Mut cells. If this hypothesis holds, one would also predict intracellular glucose accumulation when the mutants are exposed to insulin since the glucose transporter and glycogen synthetic systems of these cells remained activated in the presence as well as in the absence of insulin. Consistently, Glu-6-P, the major intracellular form of free glucose, raised significantly when the Mut myotubes were exposed to insulin while decreasing in the wild-type and the untransfected cells.
A variety of mutations have been found or generated in the regulatory
region of the insulin receptor (4, 5). To our knowledge, however, the
Arg1152Gln substitution is the first defect shown to
result in a net gain of function by the receptor, i.e. the
receptor acquires the ability to signal a novel effect (inhibition of
PDH activity in response to insulin). This gain of function appears to
be unique to the mutation rather than due to the chronic signaling by
the constitutively active receptor since chronic exposure of WT cells to insulin does not allow a subsequent acute stimulation by insulin to
inhibit PDH activity (data not shown). Interestingly, in the L6 muscle
cells, IR1152 is capable of transducing an insulin signal
on glucose oxidation but not on glucose storage. One might argue,
therefore, that divergent signaling pathways control the two major
pathways of glucose metabolism in the L6 cells so that each of the two
is differentially affected by the mutant receptor activity. Due to both
redundance and/or complementation in signaling pathways (13, 33), the
precise sequence of molecular events involved in insulin regulation of glucose storage and oxidation in muscle is still unclear. There is
evidence, however, suggesting that activation of PKC is required for
normal insulin stimulation of PDH activity (29). As previously shown,
the mutation region in IR1152 is crucial for allowing PKC
to phosphorylate the receptor (10). We also have preliminary evidence
that IR1152-associated PKC as well as
IR1152-associated PKC activity decrease below the basal
levels after insulin stimulation of the
receptor.2 We suggest, therefore, that the
inhibition of PDH activity below basal levels, occurring in the Mut
cells when exposed to insulin, might depend on the abnormal interaction
of the mutant receptor with PKC.
The relevance of these observations to IR1152 expression in skeletal muscle in vivo is also underlined by our findings on the forearm arterial-deep venous glucose differences in IR1152 patients. Similarly as in the L6 myotubes, basal glucose disposal by the forearm was >2-fold greater in the patient than in control individuals. In addition, separate studies in these patients (6) have shown that forearm glucose uptake during hyperinsulinemia was even slightly smaller than that measured in the basal state although increasing by 10-fold in control individuals. We suggest, therefore, that similar mechanisms impair glucose metabolism in skeletal muscle expressing IR1152 in vitro and in vivo.
In summary, in the present paper, we have shown that the
Arg1152Gln substitution in the insulin receptor results
in a gain of function for the receptor, i.e. the ability to
signal PDH depression in response to insulin in vitro and
in vivo. Analysis of IR1152 in skeletal muscle
suggests that insulin regulation of glucose storage and oxidation
involves divergent signaling pathways.
We are grateful to Dr. S. Gammeltoft (Bispebjerg Hospital, Copenaghen) for generously donating the WT IR cDNA and to Dr. L. Beguinot (DIBIT and Istituto di Neuroscienze e Bioimmagini H.S. Raffaele, Milan) for reviewing the manuscript. We also thank Drs. S.M. Aloj, E. Consiglio, G. Salvatore, and G.C. Vecchio (University of Naples Medical School) for continuous support and advice during the course of this work and Dr. D. Liguoro for technical help.