| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Veterans Affairs San Diego HealthCare System (9111G) and Department of Medicine, University of California-San Diego (T.P.C., L.C., S.M., R.R.H.), La Jolla, California 92093; and Aventis Pharm (G.S.), D-65926 Frankfurt, Germany
Address all correspondence and requests for reprints to: T. P. Ciaraldi, Ph.D., Department of Medicine (9111G), University of California-San Diego, La Jolla, California 92093. E-mail: tciaraldi{at}ucsd.edu
Abstract
The aim of this study was to determine whether the long-acting insulin analog, insulin glargine, behaves like human insulin for metabolic and mitogenic responses in differentiated cultured human skeletal muscle cells from nondiabetic and diabetic subjects. Human insulin and insulin glargine were equipotent in their ability to compete for [125I]insulin binding. Insulin glargine displaced [125I]IGF-I from the IGF-I-binding site with approximately 0.5% the potency of IGF-I. In nondiabetic muscle cells, all three ligands stimulated glucose uptake similarly, whereas the sensitivity of glucose uptake was greatest in response to IGF-I and lower and equal for human insulin and insulin glargine. In diabetic muscle cells, the final responsiveness of glucose uptake was greatest for IGF-I and equivalent for human insulin and insulin glargine; sensitivities were the same as those for nondiabetic cells. Thymidine uptake into DNA was stimulated foremost by IGF-I, whereas human insulin and insulin glargine showed equivalent, but greatly reduced, sensitivities and potencies (<1% IGF-I). Stimulation of Akt phosphorylation was slightly more responsive to IGF-I compared with human insulin and insulin glargine, with sensitivities similar to glucose uptake stimulation. We conclude that in human skeletal muscle cells, insulin glargine is equivalent to human insulin for metabolic responses and does not display augmented mitogenic effects.
ONE IMPORTANT APPROACH, both in the study of insulin action and the development of new antidiabetic therapies, has been the use of structurally modified insulin molecules. Changes in the insulin molecule through naturally occurring mutations and chemical or molecular biological modifications can lead to changes in the pharmacokinetics (1) and bioavailability (2, 3) of the insulin analogs, which are important issues in drug design. Structurally modified insulins can also differ from the native hormone with regard to receptor binding affinity (4, 5), negative cooperativity (6), ligand internalization and recycling (7), and selectivity for biological responses (4). A prime example of this is [AspB10]insulin, which is only slightly more effective than native insulin for receptor binding and the stimulation of lipogenesis, but is 4 times more potent for mitogenic activity (4) and the activation of receptor kinase activity (8).
Another modified insulin of interest is [GlyA21,ArgB31,ArgB32]insulin (insulin glargine), which displays delayed absorption and a prolonged course of action (9), yet dissociates from the insulin receptor slightly more rapidly than native insulin (10). Insulin glargine (LANTUS; Aventis Pharm, Frankfurt, Germany) has been tested in patients with diabetes and found to be effective in a once daily dosing regimen (9). It is critical to realize that the C-terminus of the insulin B chain, where some of the modifications in insulin glargine occur, is important in determining how the insulin molecule interacts with the IGF-I receptor (11, 12). Since the IGF-I receptor can mediate the growth-promoting effects of ligands other than its cognate hormone, the possible augmented proliferative actions of structurally modified insulins must be considered in evaluating the potential therapeutic utility of these molecules. Several studies have addressed this question with regard to insulin glargine. In rat fibroblasts overexpressing the human insulin receptor, insulin and insulin glargine were found to be similar with respect to insulin receptor binding, the activation of early signaling events, and the stimulation of mitogenesis (10). Insulin glargine and insulin were also comparable in rat cardiac muscle cells, and both insulin molecules had a less potent effect on mitogenesis than IGF-I (13). However, another study found that insulin glargine had augmented affinity for the IGF-I receptor and mitogenic potency compared with native insulin (14). There are important differences between each of these studies. In rat fibroblasts (10), the insulin receptor was expressed far in excess of the endogenous IGF-I receptor, which might mask possible actions through the IGF-I receptor. The muscle cell line employed for mitogenesis studies, H9c2 cardiac myoblasts (13), lacks insulin receptors and is unlikely to be representative of skeletal muscle. The report by Kurtzhals et al. (14) used IGF-I receptors purified from transfected cells for the determination of binding affinities, and Saos/B10 human osteosarcoma cells for mitogenesis measurements. Differences in the relative expression of insulin and IGF-I receptors could complicate comparisons between the studies.
We elected to determine whether insulin glargine had any acute metabolic or chronic mitogenic actions in human skeletal muscle cells grown in culture. We have reported that differentiated muscle cell myotubes display the morphological, biochemical, and metabolic properties of skeletal muscle, including insulin responsiveness for glucose metabolism (15). Most importantly, cultured muscle cells from patients with type 2 diabetes exhibit impairments in glucose uptake and glycogen synthase that are highly reflective of defects seen in vivo (16, 17).
Materials and Methods
Study design
As skeletal muscle is a primary target for injected insulin glargine, cultured human skeletal muscle cells were employed to evaluate the metabolic and mitogenic effects of insulin formulations, as these cells possess the metabolic machinery equivalent to that in intact human skeletal muscle. Insulin and IGF-I receptor binding and various metabolic or mitogenic parameters (stimulation of glucose transport, thymidine incorporation, and intracellular phosphorylation of key proteins in the signaling pathways) were examined after exposure to insulin glargine. All studies were performed using human insulin and IGF-I as reference compounds; cells from both nondiabetic subjects and patients with type 2 diabetes were treated with the test and reference compounds and examined for metabolic and mitogenic effects.
Materials
Human biosynthetic insulin and IGF-I were obtained from Calbiochem (La Jolla, CA). Insulin glargine ([GlyA21,ArgB31,ArgB32]insulin) was provided by Aventis Pharma (Frankfurt/Main, Germany). Cell culture materials were purchased from Irvine Scientific (Irvine, CA), except for skeletal muscle basal medium, which was obtained from Clonetics Corp. (San Diego, CA). FBS was purchased from Gemini (Calabasas, CA). Roche (Indianapolis, IN) supplied BSA (Cohn fraction V). [125I]Insulin, [125I]IGF-I, 2-[1,2-3H]deoxy-D-glucose, L-[1-14C]glucose, and [3H]thymidine were purchased from NEN Life Science Products (Boston, MA). Polyclonal antibodies against phosphoSer473-Akt and phosphoMAPK were purchased from New England Biolabs, Inc. (Beverly, MA). The antirabbit IgG conjugated to horseradish peroxidase was obtained from Amersham Pharmacia Biotech (Arlington Heights, IL), and the SuperSignal chemiluminescence substrate was purchased from Pierce Chemical Co. (Rockford, IL).
Human subjects
Seventeen healthy nondiabetic subjects and 16 patients with type
2 diabetes participated in these studies. Glucose tolerance was
determined from a 75-g oral glucose tolerance test. Insulin action was
determined by a 3-h hyperinsulinemic (300
mU/m2·min euglycemic (5.0–5.5 mM)
clamp; the glucose disposal rate was measured during the last 30 min of
the clamp (18). Patients with type 2 diabetes had their
medication withheld on the morning of the biopsy. Subject
characteristics are summarized in Table 1
. The diabetic group was significantly
older and more obese than the nondiabetic group. Impairments of the
maximally insulin-stimulated glucose disposal rate confirmed the
insulin resistance of the diabetic group (Table 1). The experimental
protocol was approved by the committee on human investigation of the
University of California-San Diego. Informed written consent was
obtained from all subjects after explanation of the protocol.
|
Biopsy of the vastus lateralis muscle was performed according to
published procedures (19). Human skeletal muscle cells
were isolated and grown in culture as described in detail previously
(15). When myoblasts attained 80–90% confluence, the
growth medium was changed to 
MEM supplemented with 2% FBS and
antibiotics to induce fusion to multinucleated myotubes. The medium was
changed every other day during cell fusion. All studies were performed
on cells after one passage. Due to the limited number of cells
available from a biopsy, not all studies were performed in each
subject.
Insulin and IGF-I binding assays
Hormone binding assays were performed by the modification of a method described previously (15). Fully differentiated human skeletal muscle cells from nondiabetic subjects and patients with type 2 diabetes were washed four times with reaction buffer, then incubated with reaction buffer and [125I-Tyr A14]insulin (final concentration, 67 pM) or [125I-Tyr A14]IGF-I (final concentration, 39 pM) for 4 h at 12 C in the absence or presence of varying concentrations of unlabeled hormone. Results were calculated based on the displacement of specific insulin/IGF-I binding normalized to protein concentrations in cells from both nondiabetic subjects and patients with type 2 diabetes.
Glucose uptake assay
Glucose uptake measurements were carried out using a previously published procedure (17). Medium was added to the cells, which were derived from both nondiabetic subjects and patients with type 2 diabetes, together with peptides, and the cells were incubated for 60–90 min at 37 C in a 5% CO2 incubator before washing and performing the transport assay. An aliquot of the suspension was removed for protein analysis using the Bradford method (20). The uptake of L-glucose was used to correct each sample for the contribution of diffusion. Dose-response curves were calculated as a function of the basal (no insulin added) sample in each set of cells, either nondiabetic cells or cells from patients with type 2 diabetes. This procedure measures both the transport and phosphorylation of 2-deoxyglucose. Earlier studies have shown that transport is the rate-limiting process under these conditions.
Thymidine incorporation into DNA
A modification of the method described by Berhanu et
al. (21) was employed. Preliminary studies indicated
that prior serum starvation was not required for optimal hormone
responsiveness in human skeletal muscle cells (not shown). Confluent
cells from nondiabetic subjects and patients with type 2 diabetes were
treated in serum-free (0.1% BSA) MEM with varying concentrations of
ligands for 16 h in a 37 C, 5% CO2
incubator. The media were replaced with MEM/0.1% BSA, pH 7.4,
together with any ligands, and [3H]thymidine
(0.5µCi) was added to each well. The cells were incubated for 2
h at 37 C, then rinsed twice with 4 C PBS and washed once with
methanol, twice with 5% trichloroacetic acid (wt/vol), followed by
ethanol at 4 C. The cells were solubilized in 1 N
sodium hydroxide (NaOH) and neutralized with an equal volume of 1
N hydrochloric acid (HCl). Aliquots were removed
for scintillation counting and protein determination by the Bradford
method, using BSA as the standard. Results are presented as a
percentage of the total thymidine added incorporated into DNA for each
individual set of cells and normalized to cell protein.
Intracellular signal transduction
Cells were washed free of medium and then treated with varying concentrations of ligands for 15 min at 37 C. This period provides maximal insulin stimulation of tyrosine phosphorylation and activation of PI3K (not shown). Reactions were terminated by washing with 4 C PBS, and cells were lysed in a buffer containing 20 mM Tris-HCl, 145 mM sodium chloride (NaCl), 10% glycerol, 5 mM EDTA, 1% Triton X-100, 0.5% Nonidet P-40, 200 µM sodium orthovanadate, 200 µM phenylmethylsulfonylfluoride, 1 µM leupeptin, 1 µM pepstatin, and 10 µg/ml aprotinin, pH 7.5. After assay for protein content, extracts were stored at -70 C before further analysis. Proteins were separated by SDS-PAGE on 10% gels. Western blot analysis was performed by the method of Burnette (22) as described previously (15). The secondary antibody was antirabbit IgG conjugated with horseradish peroxidase. Proteins were visualized with the SuperSignal Western Blot Detection Kit (Pierce Chemical Co.) and exposed to autoradiograph film. The intensities of the phosphorylated bands were quantified by scanning laser densitometry using ScanAnalysis software (Biosoft, Cambridge, UK). Membranes were stripped and reprobed with antibodies against total specific protein (Akt or MAPK) to ensure equal loading of the gel and transfer to the membrane.
Statistical analysis
Data calculations and statistical analyses were preformed using the StatView SE+ program (Abacus Concepts, Berkeley, CA). All data are expressed as the mean ± SEM. For all variables, statistical significance was tested with two-tailed t test, using paired analysis when appropriate. Significance was accepted at P < 0.05. Results presented as a percentage or fold change were obtained from paired comparisons with the appropriate control for each set of cells from that single individual. Ligand sensitivities (EC50 values) were determined from log-logit transformations of individual dose-response curves. Maximal responses were determined in two ways: for glucose uptake it was the greatest activity obtained for each individual hormone for that subjects’ cells; for thymidine uptake, maximal responsiveness was calculated as both the greatest absolute response obtained in each subjects’ cells, always in response to IGF-I, or the greatest response obtained for each individual hormone.
Results
Insulin receptor binding
Conditions for the receptor binding assays (4 h at 12 C) were selected so that only cell surface binding was measured. Under these conditions, internalization and recycling of the ligand is minimal (23).
In muscle cells from nondiabetic subjects, the average concentration of
human insulin required to displace 50% of the specifically bound
insulin, an approximation of ligand affinity for the receptor, was 0.72
nM (Table 2). Insulin
glargine was indistinguishable from human insulin with regard to
displacement of labeled insulin; the competition curves were
superimposable (Fig. 1A). The average
affinity for insulin glargine (1.07 nM) did not differ from
that of insulin (P = NS).
|
|
and Table 2). Once again, insulin glargine behaved the same as human
insulin in competition for the labeled insulin; the affinities did not
differ significantly. IGF-I receptor binding
The specific binding of a tracer concentration of IGF-I to
cultured skeletal muscle cells is, on the average, nearly 5-fold
greater than that of insulin (compare Figs. 1 and 2). IGF-I affinity for its receptor is
approximately 2- to 3-fold greater than that of insulin for its
receptor (Table 2). In nondiabetic cells, both insulin glargine and
insulin displaced bound IGF-I with very low affinity (Fig. 2). Human
insulin and insulin glargine competed for the IGF-I receptor with
similar affinities (P = NS). Only at the highest level
tested did insulin glargine compete for IGF-I binding significantly
better than human insulin (Fig. 2). Cross-talk of insulin glargine for
the IGF-I receptor was approximately 0.25% that of IGF-I. A 50%
effective binding value could not be calculated for human
insulin, as 50% displacement of IGF-I binding was not attained over
the concentration range tested.
|
). It was only at the highest level tested in
these studies (200 nM) that insulin glargine displaced more
IGF-I than human insulin. Despite this difference, insulin glargine has
less than 0.1% of the affinity of the native ligand for the IGF-I
receptor (Table 2). Glucose uptake stimulation
Stimulation of glucose uptake was measured in muscle cells from
nondiabetic subjects and patients with type 2 diabetes. There was
considerable subject-subject variation in the nondiabetic subject
group: basal activity was 18.8 ± 3.5 pmol/mg protein·min,
whereas maximally stimulated activity was 30.1 ± 3.9. Normalizing
the results against the basal activity for each individual reduced the
variability (Fig. 3). Human insulin,
insulin glargine, and IGF-I all stimulated glucose uptake in a
dose-dependent manner (Fig. 3). In muscle cells from nondiabetic
subjects there was no difference (by paired analysis, P
> 0.4) in the maximal responsiveness to any of the ligands (Fig. 3A).
Muscle cells were similarly sensitive to human insulin and insulin
glargine (Table 3), with
EC50 values close to their binding affinities for
the insulin receptor, suggesting that they are functioning through that
receptor. Muscle cells were more sensitive to IGF-I (P
< 0.05), reflecting the presence of a higher number of receptors with
slightly greater affinity. The high sensitivity to IGF-I indicates that
IGF-I is acting through its own receptor and not by cross-talk through
the insulin receptor.
|
|
). The results indicate that type 2 diabetic muscle
cells display greater resistance to human insulin than IGF-I, and that
insulin glargine was unable to overcome this insulin resistance. Thymidine uptake
To investigate more chronic responses to human insulin, insulin glargine, and IGF-I, [3H]thymidine uptake into DNA was measured. As this assay was performed on fused/differentiated cells, further cell division was not possible, and this activity does not represent a true proliferative response. However, prolonged exposure to the peptides (16–18 h) was necessary to obtain the maximal response, and it does indicate the effects of extended treatment.
In muscle from nondiabetic subjects, all three peptides caused a
dose-dependent increase in thymidine uptake (Fig. 4A). The dose-response curves for human
insulin and insulin glargine were shifted far to the right of that for
IGF-I. At the highest concentration tested, neither peptide was as
effective as IGF-I (Fig. 4A). The relationship between the curves
appears similar to the abilities of insulin and insulin glargine to
displace IGF-I from its receptor (Tables 2 and 3), suggesting that
much, if not all, of the effect of these ligands to stimulate thymidine
uptake is occurring as a result of cross-talk through the IGF-I
receptor. Interestingly, although insulin glargine was slightly more
effective than insulin in occupying the IGF-I receptor (Table 2), it
had less of an effect on thymidine uptake. The sensitivities of human
insulin and insulin glargine for thymidine uptake were comparable when
calculated against the maximal possible response in each set of cells
(due to IGF-I; Table 3), and in terms of the highest response seen for
each individual hormone (not shown).
|
). The difference was most notable for IGF-I and
may represent a generalized hormone resistance, unlike the
insulin-specific resistance of the glucose transport response. The
sensitivity to IGF-I was unaltered in diabetes (Table 3). Responses to
human insulin and insulin glargine exceeded 50% of the IGF-I effect in
diabetic cells; therefore, sensitivities could be calculated for these
peptides as well (Table 3). From consideration of the absolute values
it appears that this response to human insulin and insulin glargine is
generally unaltered in diabetes. Intracellular signaling
As insulin stimulation of early intracellular signaling events is
transient, cells were treated for 15 min to attain maximum effects on
the phosphorylation of key proteins and enzymes involved in signaling.
One key step is insulin-stimulated serine phosphorylation of the
serine/threonine kinase Akt, which correlates with activation of the
kinase (24). As phosphoS-Akt is often undetectable
in the absence of a ligand (Fig. 5A),
results are expressed as a percentage of the maximal Akt
phosphorylation attained in each individual set of cells. In that way
the relative effects of human insulin, insulin glargine, and IGF-I can
be compared. All three ligands stimulated serine phosphorylation in a
dose-dependent manner (Fig. 5A). In nondiabetic cells, final
responsiveness was only slightly (P = NS) greater for
IGF-I (Fig. 5B). There was no significant difference between the
responsiveness to human insulin or insulin glargine. Sensitivities to
IGF-I (EC50, 
0.3 nM) and
insulin (2 nM) were similar to those for
glucose uptake stimulation (Table 3). The curve for insulin glargine
was shifted somewhat to the right of that for insulin
(EC50, 5 nM); the
physiological significance of this small difference is uncertain.
Generally similar behavior was observed in cells from diabetic subjects
(Fig. 5B). IGF-I responsiveness was significantly (P <
0.05) greater than that for human insulin or insulin glargine, which
were comparable, whereas the sensitivity was essentially the same as
that in nondiabetic cells (P = NS).
|
), and the maximal hormone effect was
increased by an average of 100% (Fig. 6B). Increases of the same
magnitude have been seen for MAPK phosphorylation in skeletal muscles
after insulin treatment (27, 28). We used this protocol
because it more closely resembled the in vivo situation.
Once again, in nondiabetic muscle cells the greatest final response was
attained after IGF-I stimulation (P < 0.05
vs. human insulin and insulin glargine). Human insulin and
insulin glargine were equipotent (Fig. 6B), similar to the behavior
observed for stimulation of [3H]thymidine
uptake (Fig. 4). In diabetic muscle cells, the IGF-I response was
relatively lower, much as what was seen for the stimulation of
thymidine uptake (Fig. 4B).
|
A number of approaches have been used to study the relationship between the structure and function of insulin. Molecular biological techniques have generated multiple insulin analogs with both enhanced (4) and reduced (6, 29, 30, 31) biological potencies. Most importantly, modified insulins have been identified with accelerated (3) and prolonged (9) action profiles. The advantages of a long-acting insulin analog include reduced frequency of treatment and a flatter, more physiological action profile, which could avoid nocturnal hypoglycemia (1). Changes in the amino acid sequence at the C-terminus of the insulin B chain have produced insulins with altered physical properties that are absorbed more slowly at physiological pH (2). An example of such an insulin is [GlyA21,ArgB31,ArgB32]insulin (insulin glargine), which displays a delayed onset and longer duration of action compared with native insulin (9). It is important to realize that changes in the region of the insulin molecule (C-terminus) that produce these desired properties also alter the affinity of the molecule for the IGF-I receptor (11, 12, 32), giving rise to the possibility of increased proliferative actions, even tumorigenesis. Thus, longer acting insulin analogs must be closely monitored for their mitogenic potential. Several such studies have been performed for insulin glargine, with conflicting results. In several systems, insulin glargine was equipotent with insulin for metabolic and mitogenic responses (10, 13), whereas others reported an augmented mitogenic efficacy (14). One explanation for these differences could be the varying insulin and/or IGF-I receptor densities in the cell systems studied. Species differences may also play a role. For example, the in vivo hypoglycemic potency of another insulin analog, [AlaB5]insulin, differs among rats, mice, and rabbits (33).
In the current report we have employed a system of human skeletal muscle cells in culture to compare and contrast insulin glargine with human insulin and IGF-I with regard to their affinity for the insulin and IGF-I receptors as well as their stimulation of representative metabolic and mitogenic responses and signaling events proximal to these responses. One advantage of this system is that the relative expression of insulin and IGF-I receptors mirrors that seen in skeletal muscle, the major insulin target tissue.
Insulin glargine displaced human insulin from its receptor with an efficiency indistinguishable from that of native insulin, a result in agreement with earlier reports in other systems (10, 13, 14). We selected conditions for the binding assay that would detect binding to cell surface receptors (23). Other studies employed either solubilized receptors (14) or higher temperatures where additional events such as ligand internalization and retroendocytosis could complicate analysis of the results. The comparability of insulin and insulin glargine under these conditions suggests that insulin glargine has the same intracellular itinerary as insulin. This comparability was also seen in muscle cells from patients with type 2 diabetes, which retain defects in insulin-stimulated glucose metabolism (16, 17).
Comparability of human insulin and insulin glargine was also seen for the stimulation of glucose uptake in muscle cells with regard to both final responsiveness and sensitivity. Sensitivities for this metabolic response were in a range similar to those seen for insulin receptor occupancy, indicating that glucose uptake is mediated through the insulin receptor. The same conclusion was drawn for the stimulation of transport in several other cell types (13, 14). Thus, human insulin and insulin glargine appear to act through common mechanisms to couple receptor binding to metabolic responses. Human muscle cells were also highly sensitive to glucose uptake stimulation by IGF-I. The agreement between sensitivities for transport stimulation and receptor binding indicate that, in human skeletal muscle cells at least, IGF-I is able to stimulate through its own receptor metabolic responses independent of the insulin receptor. Just as binding affinities are unaltered by the presence of diabetes, sensitivities for this response are not affected by the diabetic status when cells are maintained in an euglycemic, normoinsulinemic environment. However, maximal responsiveness is reduced in diabetic muscle cell cultures, reflecting the donor’s in vivo insulin resistance.
Specific binding of a tracer concentration of IGF-I to human skeletal muscle cells is approximately 5-fold that of insulin in the same cells. A similar (3- to 4-fold) preponderance of IGF-I binding was observed in skeletal muscle tissue (34), indicating that the relative expression of IGF-I and insulin receptors in cultured human skeletal muscle cells is reflective of that in muscle. A portion of IGF-I binding may also be due to IGF-binding proteins, which are known to be produced by cultured muscle cells (35, 36). As insulin has not been shown to displace IGF-I from IGF-binding proteins (37), our data should represent the affinities of human insulin and insulin glargine for the IGF-I receptor. Both of these ligands are able to compete for IGF-I binding with less than 1% the potency of IGF-I, the same relative potency as has been reported previously (38, 39). A slightly higher affinity of insulin glargine, compared with insulin, for the IGF-I receptor was also noted by Bahr et al. (13). A considerably higher relative affinity (6-fold) was observed using solubilized human IGF-I receptors (14). The nature of this preparation may account for the difference from the results in intact cells. Although insulin glargine appears to have a somewhat higher affinity for the IGF-I receptor than human insulin, the difference is seen only at very high, nonphysiological concentrations. The affinity of insulin glargine for the IGF-I receptor is only 0.1–0.25% that of IGF-I, and is, therefore, expected to be of limited significance.
Where the interaction of human insulin or insulin glargine with the IGF-I receptor appears to be important is in the control of thymidine uptake into DNA. Both human insulin and insulin glargine stimulated thymidine uptake in muscle cells with greatly reduced sensitivities and potencies compared with IGF-I. The sensitivity of muscle cells to thymidine uptake in response to these ligands is of the same order of magnitude as the displacement of IGF-I binding and is far removed from affinities for the insulin receptor or sensitivity for the stimulation of glucose uptake. In muscle cells, the closest agreement is between the sensitivity for thymidine uptake and the affinity for the IGF-I receptor. This relationship applies to human insulin and insulin glargine, as well as IGF-I. Thus, it is highly likely that most, if not all, of the ability of human insulin and insulin glargine to stimulate thymidine uptake is mediated through the IGF-I receptor. In terms of thymidine uptake, human insulin and insulin glargine are comparable, as reported previously in rat fibroblasts overexpressing the insulin receptor (10) and cultured myoblasts (13). The augmented, relative to insulin, mitogenic potency of insulin glargine was seen only in osteosarcoma cells (14). In other, more classic insulin target tissues, it appears that, with regard to the activity represented by thymidine uptake and how it might relate to potential proliferative responses, insulin glargine is even less effective than human insulin.
Although the current results strongly suggest that human insulin and insulin glargine exert their mitogenic actions primarily through the IGF-I receptor, there is evidence that insulin can also activate mitogenic responses through the insulin receptor. The best example of this is in the LB strain of thymocytes, which lack IGF-I receptors and respond mitogenically to physiological insulin levels (40). In contrast, in muscle cells, specific IGF-I binding exceeds insulin binding and appreciable mitogenic responses occur only at supraphysiological insulin or insulin glargine levels. As described by Shymko et al. (41, 42) and confirmed by Hansen et al. (43), the primary determinant of selectivity for mitogenic vs. metabolic signaling mediated by the insulin receptor is the time of receptor occupancy; slower dissociating insulin analogs are more mitogenic than the native hormone (4, 40, 43). With this analysis, the best correlation of relative mitogenic potencies of insulin analogs is the dissociation rate from the insulin receptor (40, 41). Kurtzhals et al. (14) reported little relationship between rates of dissociation from the insulin receptor and mitogenesis; the best correlation was with affinity for the IGF-I receptor, suggesting a different manner of signaling in this context. Insulin glargine has been shown in most reports to dissociate slightly faster than insulin (10), in keeping with our observation of near equivalence with human insulin for mitogenesis. In contrast, AspB10 insulin displays slower dissociation, prolonged phosphorylation of intracellular proteins, and an augmented mitogenic potency (4, 8), all consistent with the relationship between time of receptor occupancy and selectivity for final responses (41). The prolonged action of insulin glargine on glucose metabolism is most likely to result from the altered pharmacokinetics (9) and not extended receptor occupancy.
Insulin and IGF-I binding to their respective receptors activates a number of signaling cascades. The pathway involving shc-Grb2/SOS-ras, which leads to the phosphorylation and activation of MAPK, has been implicated in cell growth, including stimulation of thymidine uptake (44), with little or no involvement in metabolic responses (45). The phosphorylation and activation of Akt via PI3K has been implicated in a number of metabolic responses to growth factors, such as the stimulation of glucose uptake (46, 47). The relative behaviors of human insulin, insulin glargine, and IGF-I for selected proximal events in growth factor signaling cascades are very similar to the behaviors for metabolic and mitogenic responses. Earlier work showed that insulin and insulin glargine were equivalent for insulin receptor phosphorylation and tyrosine phosphorylation of insulin receptor substrate-1 and Shc (10). That report focused on the kinetics of phosphorylation, whereas we looked at a single time point. In the current work we found that the three ligands stimulated Akt phosphorylation with sensitivities similar to those for the stimulation of glucose uptake. The maximal responsiveness of both glucose uptake and Akt phosphorylation to all three ligands was comparable in muscle cells from nondiabetic subjects, whereas in diabetic muscle cells the final IGF-I stimulation was greatest for both responses. This similar pattern of behavior for Akt phosphorylation and glucose uptake bolsters the body of evidence that places Akt in the cascade leading to glucose transport (46, 47). Human insulin and insulin glargine are also comparable for their actions on mitogenesis and MAPK phosphorylation, with IGF-I being more potent in nondiabetic cells.
The current results indicate that in a cell system representative of the relative insulin and IGF-I receptor expression observed in intact skeletal muscle, human insulin and insulin glargine are fully comparable for binding to insulin and IGF-I receptors, stimulation of Akt and MAPK phosphorylation, stimulation of glucose uptake, and mitogenesis. The latter response occurs via the IGF-I receptor, whereas in muscle cells IGF-I generates all of these responses through its own receptor. The augmented mitogenic action of insulin glargine is not observed in the major therapeutic target tissue for this insulin analog. In human skeletal muscle, significant mitogenic responses to human insulin and insulin glargine occur at levels far higher than those likely to be achieved in the circulation.
Acknowledgments
Footnotes
This work was supported by a grant from the Medical Research Service, Department of Veterans Affairs and V.A. San Diego HealthCare System, funds from Hoechst Marion Roussel, Inc., and Grant MO1-RR-00827 from the General Clinical Research Branch, Division of Research Resources, NIH.
Received June 22, 2001.
Accepted September 5, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  A. Shukla, J. Grisouard, V. Ehemann, A. Hermani, H. Enzmann, and D. Mayer Analysis of signaling pathways related to cell proliferation stimulated by insulin analogs in human mammary epithelial cell lines Endocr. Relat. Cancer, June 1, 2009; 16(2): 429 - 441. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Erbel, C. Reers, V. W. Eckstein, J. Kleeff, M. W. Buchler, P. P. Nawroth, and R. A. Ritzel Proliferation of Colo-357 Pancreatic Carcinoma Cells and Survival of Patients With Pancreatic Carcinoma Are Not Altered by Insulin Glargine Diabetes Care, June 1, 2008; 31(6): 1105 - 1111. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U Berg, T Gustafsson, C J Sundberg, L Kaijser, C Carlsson-Skwirut, and P Bang Interstitial IGF-I in exercising skeletal muscle in women Eur. J. Endocrinol., October 1, 2007; 157(4): 427 - 435. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Jovanovic and D. J. Pettitt Treatment With Insulin and Its Analogs in Pregnancies Complicated by Diabetes Diabetes Care, July 1, 2007; 30(Supplement_2): S220 - S224. [Full Text] [PDF]
|
|
|
|
 |
|
 T. P. Ciaraldi, S. A. Phillips, L. Carter, V. Aroda, S. Mudaliar, and R. R. Henry Effects of the Rapid-Acting Insulin Analog Glulisine on Cultured Human Skeletal Muscle Cells: Comparisons with Insulin and Insulin-Like Growth Factor I J. Clin. Endocrinol. Metab., October 1, 2005; 90(10): 5551 - 5558. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Mohn, M. Marcovecchio, F. Chiarelli, M. C. Haffner, M. P. Kufner, B. O. Boehm, and I. B. Hirsch Insulin Analogues N. Engl. J. Med., April 28, 2005; 352(17): 1822 - 1824. [Full Text] [PDF]
|
|
|
|
 |
|
 S. I. Chisalita and H. J. Arnqvist Insulin-like growth factor I receptors are more abundant than insulin receptors in human micro- and macrovascular endothelial cells Am J Physiol Endocrinol Metab, June 1, 2004; 286(6): E896 - E901. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Castillo, M. Codina, M. L. Martinez, I. Navarro, and J. Gutierrez Metabolic and mitogenic effects of IGF-I and insulin on muscle cells of rainbow trout Am J Physiol Regulatory Integrative Comp Physiol, May 1, 2004; 286(5): R935 - R941. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Westerbacka, R. Bergholm, M. Tiikkainen, and H. Yki-Jarvinen Glargine and Regular Human Insulin Similarly Acutely Enhance Endothelium-Dependent Vasodilatation in Normal Subjects Arterioscler. Thromb. Vasc. Biol., February 1, 2004; 24(2): 320 - 324. [Abstract] [Full Text]
|
|
|
|
|
|
S. Vehkavaara and H. Yki-Jarvinen 3.5 Years of Insulin Therapy With Insulin Glargine Improves In Vivo Endothelial Function in Type 2 Diabetes Arterioscler. Thromb. Vasc. Biol., February 1, 2004; 24(2): 325 - 330. [Abstract] [Full Text]
|
|
|
|
 |
|
 B. Ahmad Review: Pharmacology of insulin The British Journal of Diabetes & Vascular Disease, January 1, 2004; 4(1): 10 - 14. [Abstract] [PDF]
|
|
|
|
 |
|
 W. Niu, C. Huang, Z. Nawaz, M. Levy, R. Somwar, D. Li, P. J. Bilan, and A. Klip Maturation of the Regulation of GLUT4 Activity by p38 MAPK during L6 Cell Myogenesis J. Biol. Chem., May 9, 2003; 278(20): 17953 - 17962. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
