Micro- and macroangiopathy are major causes of morbidity and mortality in patients with diabetes. Our aim was to characterize IGF-I receptor (IGF-IR) and insulin receptor (IR) in human micro- and macrovascular endothelial cells. Cultured human dermal microvascular endothelial cells (HMVEC) and human aortic endothelial cells (HAEC) were used. Gene expression was measured by quantitative real-time RT-PCR and receptor protein by ligand-binding assay. Phosphorylation of IGF-IR β-subunit was analyzed by immunoprecipitation and Western blot. Glucose metabolism and DNA synthesis was assessed using [3H]glucose and [3H]thymidine incorporation, respectively. We detected gene expression of IGF-IR and IR in HAEC and HMVEC. IGF-IR gene expression was severalfold higher than that of IR. The specific binding of 125I-IGF-I was higher than that of 125I-insulin in HAEC and HMVEC. Insulin and the new, long-acting insulin analog glargine interacted with the IGF-IR with thousand- and hundred-fold less potency than IGF-I itself. Phosphorylation of the IGF-IR β-subunit was shown in HAEC for IGF-I (10−8 M) and insulin (10−6 M) and in HMVEC for IGF-I and glargine (10−8 M, 10−6 M). IGF-I 10−7 M stimulated incorporation of [3H]thymidine into DNA, and 10−9–10−7 M also the incorporation of [3H]glucose in HMVEC, whereas glargine and insulin had no significant effects at 10−9–10−7 M. Human micro- and macrovascular endothelial cells express more IGF-IR than IR. IGF-I and high concentrations of glargine and insulin activates the IGF-IR. Glargine has a higher affinity than insulin for the IGF-IR but probably has no effect on DNA synthesis at concentrations reached in vivo.
- human endothelial cells
in patients with diabetes mellitus, micro- and macroangiopathy are major causes of morbidity and mortality (17, 41). Situated at the interface between the blood stream and the vessel wall, the vascular endothelium has a unique position, being involved in regulation of metabolic, hemostatic, and immunologic processes (9). Endothelial dysfunction is considered to play an important role in pathogenesis of diabetic micro- and macroangiopathy, as well as in atherosclerosis (6, 34, 40).
The insulin receptor (IR) and the insulin-like growth factor I receptor (IGF-IR) are homologs sharing >50% of their amino acid sequence, and they have 84% homology in the β-subunit tyrosine kinase domains (42, 43). Binding of insulin and IGF-I to the extracellular α-subunits and autophosphorylation of the β-subunit are the first steps in the activation of IR and IGF-IR (32). This is followed by activation of a complex cascade of signaling pathways through which metabolic and growth effects are initiated (10, 36). At high concentrations, insulin and IGF-I can cross-react with each other's receptors (31). In tissues coexpressing IR and IGF-IR, hybrid receptors composed of an IR αβ-heterodimer and an IGF-IR αβ-heterodimer have been reported (26, 30). If IR and/or IGF-IR are expressed in human endothelium, a direct effect of insulin and/or IGF-I could have a critical impact on vascular function.
There are few studies regarding IGF-IR and IR in human endothelial cells. In rat and bovine micro- and macrovascular endothelial cells, both IR and IGF-IR have been shown (19, 21, 22). Specific binding of IGF-I and insulin has been demonstrated in human umbilical vein endothelial cells (4, 5, 45). In kidney (29) and retinal endothelial cells (39), IGF-IR have been reported. There is a lack of studies regarding effects of insulin and IGF-I on glucose metabolism and DNA synthesis in human endothelial cells. An effect of IGF-I on [3H]thymidine incorporation into DNA has been reported in human retinal endothelial cells (14). The aim of our study was to characterize IR and IGF-IR in human micro- and macrovascular endothelial cells with regard to gene expression, ligand binding, receptor activation, and the biological effect of insulin, glargine, and IGF-I. We used human dermal microvascular endothelial cells (HMVEC) as microvascular cells, and human aortic endothelial cells (HAEC) as macrovascular endothelial cells.
MATERIALS AND METHODS
Culture of cells.
HMVEC and HAEC were purchased from Clonetics (San Diego, CA) and grown according to the manufacturer's instructions. The cells were positive for von Willebrand factor VIII and acetylated LDL and negative for smooth muscle α-actin (Clonetics). In addition to the above staining, HMVEC also tested positive for platelet endothelial cell adhesion molecule. All experiments were performed in triplicate with cells in passages 5–8 in confluent cultures. All experiments were performed three to five times as mentioned in the figure legends.
Quantitative real-time RT-PCR.
RNA was extracted using the RNeasy Mini Kit (Qiagen, Hilden, Germany). From 1 μg of RNA, first-strand complementary DNA (cDNA) was transcribed using a commercial kit (Invitrogen Life Technologies, Stockholm, Sweden). The expression of IR and IGF-IR was estimated by real-time quantitative PCR action assay using the ABI PRISM 7700 Sequence Detection System (PE Applied Biosystems, Stockholm, Sweden). The oligonucleotides were purchased from SGS (Scandinavian Gene Synthesis, Koping, Sweden). For human IR (hIR), we used as forward primer 5′-AGGAGCCCAATGGTCTGA-3′, as reverse primer 5′-GAGACGCAGAGATGCAGC-3′, and as probe 5-(FAM) ACCATATCGCCGATAACTCACTTCATACAGT (TAMRA)-3′. For human IGF-IR (hIGF-IR), we used as forward primer 5′-CGATGTGTGAGAAGACCACCA-3′, as reverse primer 5′-ACATTTTCTGGCAGCGGTTT, and as probe 5′-(FAM) CAATGAGTACAACTACCGCTGCTGGACCAT (TAMRA)-3′. The primers and probe for the GAPDH gene were obtained from PE Applied Biosystems, with the sequence for forward primer 5′-GAAGGTGGAGGTCGGAGTC-3′, for reverse primer 5′-GAAGATGGTGATGGGATTTC, and probe 5′-(VIC) CAAGCTTCCCGTTCTCAGCC (TAMRA)-3′.
The real-time RT-PCR reaction was run in the 7700 Prism Sequence Detector System. A final volume of 25 μl contained 12 μl of TaqMan Universal PCR Master Mix (PE Applied Biosystems), 3 μl of cDNA of reverse-transcribed RNA, 300 nM primers, a 25 nM probe for hIR, and a 50 nM probe for hIGF-IR, respectively. After 2 min at 50°C and 10 min at 95C, the reaction ran for 40 cycles consisting of a denaturation or melting step at 95C for 15 s followed by an annealing/extension step at 60C for 1 min. The detection of the PCR products was allowed through the combination of 5′-3′ nuclease activity of AmpliTaq Gold DNA Polymerase ROX by the release of a fluorescent reporter FAM for hIR and hIGF-IR, respectively, and VIC for GAPDH probe oligonucleotides during the RT-PCR reaction. The fluorescence was measured at each cycle. The data were analyzed using Sequence Detector version 1.7 (PE Applied Biosystems). The relative amount of transcripts, measured during the exponential phase of reaction, was determined by the comparative CT methods (Bulletin 2, PE Applied Biosystems).
A total of 40,000 of either HMVEC or HAEC were seeded and cultured in six-well plates (Clonetics). Confluent cells were incubated for 4 h at 4°C in HEPES binding buffer [pH 7. 8 with the following composition (in mmol/l): 100 HEPES, 120 NaCl, 5 KCl, 1.2 MgS04, and 8 glucose and 0.1% bovine serum albumin] with the addition of 25,000 cpm 125I-IGF-I (5.7 × 10−l2 M) or mono-125I- (Tyr A14)-human insulin (5.7 × 10−l2 M; Amersham Pharmacia Biotechnology, Buckinghamshire, UK), and unlabeled polypeptides at indicated concentration. The cells were washed three times with ice-cold PBS and then solubilized in 0.1% SDS. The radioactivity was measured in a gamma counter (LKB Wallac, Turku, Finland). Nonspecific binding of 125I-insulin or 125I-IGF-I was defined as binding in the presence of 10−5 M unlabeled insulin or 10−6 M IGF-I, respectively, and was subtracted from total binding to yield specific binding. To calculate EC50 values for insulin and glargine, it was assumed that, at a concentration of 10−4 M, 125I-IGF-I was fully displaced by these two polypeptides. Data were analyzed using the Prism program (GraphPad Software, San Diego, CA). To estimate EC50 values, the ligand-binding data for 125I-IGF-I and 125I-insulin were fit to a one-site competition equation and a two-site competition equation by use of a Marquardt-Levenberg nonlinear least squares algorithm.
Immunoprecipitation of the IGF-IR and IR.
To determine the tyrosine phosphorylation state of IGF-IR and IR β-subunit, subconfluent HMVEC and HAEC cells were cultured in 75-cm2 flasks. They were serum starved for 24 h before the experiments, and two flasks were used per experimental condition. The cells were then washed in cold F-12-BSA medium (1 mg BSA/ml F-12) and incubated for 30 min on ice with 50 μM Na3VO4 solution diluted in F-12-BSA medium. The cultures were incubated in warm (37°C) F-12-BSA medium (1 mg BSA/ml F-12) at 37C for 10 min with IGF-I, insulin, or glargine added as indicated in Fig. 4. After the incubation, the cells were lysed for 30 min on ice with a lysis buffer [containing 20 mM Tris (pH 7.5), 150 mM NaCl, 5 mM EDTA, 0.5% sodium deoxycholate, 0.5% Triton X-100, 1 mM Na3VO4, 1.5 μg/ml aprotinin, 1.5 μg/ml leupeptin, and 1 mM PMSF]. Cell lysates were centrifuged at 4C for 15 min at 20,000 g. The supernatant was transferred into new tubes and stored at −70C. Total protein content was measured by the bicinchoninic acid method (Pierce, Rockford, IL) to adjust the amount of protein used for subsequent analysis. To immunoprecipitate the IGF-IR or IR, we incubated cell lysates that contained 0.5–1 mg of total proteins with 2.5 μl of anti-IGF-I β-subunit receptor antibody, rabbit polyclonal IgG (C-20), or anti-IR β-subunit receptor antibody, rabbit polyclonal IgG (C-19), respectively (Santa Cruz Biotechnology), and 50 μl of protein A-Sepharose (Pharmacia-Upjohn, Uppsala, Sweden) were added and samples shaken gently at 4°C overnight. The immunoprecipitates were washed three times with ice-cold lysis buffer and diluted in 50 μl of 2× Laemmli sample buffer (0.125 M Trisma base, 4% SDS, 10% glycerol, 0.02% bromphenol, 4% β-mercaptoethanol, pH 6.8).
Western blot analysis.
Immunoprecipited samples were boiled for 2 min. After centrifugation, proteins in the supernatant were separated on a 7.5% SDS-PAGE gel. The separated proteins were electrotransferred onto a polyvinylidene difluoride membrane and blocked overnight with blocking buffer (0.1% Tween 20, TBS). The membrane was immunoblotted using 1:1,000 dilution of the anti-phosphotyrosine antibody (PY20) (Santa Cruz Biotechnology) for 2 h at room temperature. The proteins were visualized using a specific secondary horseradish peroxidase-linked anti-mouse monoclonal IgG antibody (Amersham Life Science, Uppsala, Sweden) followed by an enhanced chemiluminescence detection system (ECL detection). Autoradiographs were obtained by exposure to a Hyperfilm ECL (Amersham Life Science).
The membranes were stripped by heating at 60C for 30 min in stripping buffer (10% SDS, 100 mM β-mercaptoethanol, 62.5 mM Tris·HCl). They were blotted again with the same anti-IGF-IR antibody or anti-IR antibody as was used for immunoprecipitation, and the proteins were visualized as described above.
[3H]thymidine incorporation into DNA.
DNA synthesis was quantified by measuring [3H]thymidine incorporation into DNA in HMVEC according to a modified method of Nilsson and Thyberg (27). The cells were grown in 24-wells plates, serum starved for 24 h, and then incubated with 1 μCi/ml [3H]thymidine (Amersham Pharmacia Biotechnology) with and without polypeptides at indicated concentrations. DNA was precipitated with 5% cold trichloroacetic acid and then solubilized in 0.5 ml of 0.1 mol/l KOH. Part of the solution, 0.4 ml, was added to 4 ml of scintillation solution, and the radioactivity was measured in a liquid scintillation counter (Rackbeta 1217, LKB Wallac). The data were expressed as percent increase in [3H]thymidine incorporation of control cells (100%).
Confluent HMVEC were growth in six-well plates and starved in serum-free medium for 24 h before the experiment. The cells were then incubated at 37°C for 2 h with the addition of 1 μCi/ml [3H]glucose (Amersham Pharmacia Biotechnology) and in the absence or presence of insulin, glargine, and IGF-I at indicated concentrations. Cells were rinsed three times with PBS and lysed with 0.5 ml of 0.1% SDS, and 0.4 ml of the solubilized cells was added to 4 ml of scintillation solution, and the radioactivity was measured. The data were expressed as percent increase of control cells (100%).
Values are given as means ± SE. Statistical comparisons were made with the SPSS program (SPSS, Chicago, IL) by one-way ANOVA. A P value of <0.05 was considered statistically significant.
Quantitative real-time RT-PCR.
Measured by quantitative real-time PCR, gene expression of both IGF-I and IR was demonstrated in HAEC and HMVEC. The relative amount of IGF-IR to IR mRNA expression in cultured HMVEC and HAEC was about five and eight times higher, respectively (P < 0.001; Fig. 1).
IGF-I and IR protein was assessed by ligand binding using 125I-IGF-I and 125I-insulin (Fig. 2). In HMVEC, the specific binding of 125I-IGF-I was 1.6 ± 0.2% of total 125I-IGF-I added (Fig. 2). Tested in one-site or two-site competition equations, the 125I-IGF-I binding was found to fit a one-site binding model. The concentration needed to give half-maximal displacement (EC50) of labeled 125I-IGF-I was 5.7 × 10−10 M for unlabeled IGF-I, 2.5 × 10−8 M for the insulin analog glargine, and 2.2 × 10−7 M for insulin (Fig. 3A). Glargine and insulin did not fully displace 125I-IGF-I to the same level that IGF-I did. The specific binding of 125I-IGF-I in HAEC was 1.9 ± 0.1% of total 125I-IGF-I added (Fig. 2). EC50 was 4.3 × 10−10 M for unlabeled IGF-I, 9.9 × 10−8 M for glargine, and 5.8 × 10−6 M for insulin (Fig. 3B). The results were similar to those for HMVEC with the same order of potency, insulin < glargine < IGF-I, to displace 125I-IGF-I. Insulin was thus approximately a thousandfold less potent and glargine a hundredfold less potent than IGF-I to displace 125I-IGF-I from its receptor.
The specific binding of 125I-insulin was 0.5 ± 0.2% in HMVEC and 0.2 ± 0.04% in HAEC (Fig. 2). The total specific binding of 125I-insulin was thus much lower than the specific binding of 125I-IGF-I for both HAEC and HMVEC (P < 0.001; Fig. 2). Even though the specific binding of 125I-insulin was very low, it was possible to calculate approximate EC50 values. The binding of 125I-insulin to HMVEC and HAEC was tested in one-site and two-site competition equations and was found to fit a two-site binding model (Fig. 3, C and D) with a high-affinity and a low-affinity site. For displacement of 125I-insulin by unlabeled insulin and glargine, the EC50 for the high-affinity site was 10−14 to 10−12 M. For IGF-I, no low-affinity site was found in either HMVEC or HAEC. A low-affinity site, 10−8 to 10−6 M, was obtained for unlabeled insulin and glargine and also for IGF-I (Fig. 3, C and D).
Immunoprecipitation and Western blot analysis.
Immunoprecipitation and Western blot analysis were used to estimate the effect of IGF-I, insulin, and glargine on phosphorylation of IGF-IR β-subunit (Fig. 4, A–C). Cells were exposed to IGF-I, insulin, or glargine for 10 s. In both HAEC and HMVEC, we found a band at 97 kDa, a position corresponding to IGF-IR β-subunit (Fig. 4, A–C, bottom). IGF-I at 10−8 M was able to phosphorylate its own receptor in HMVEC and HAEC (Fig. 4, A and C, top). In a very high concentration (10−6 M), insulin was able to phosphorylate the IGF-IR β-subunit in HAEC, whereas no clear effect was obtained at an insulin concentration of 10−8 M (Fig. 4C, top). In separate experiments using HMVEC, we obtained phosphorylation the IGF-IR β-subunit by 10−6 M glargine and 10−6 M IGF-I and also a faint band by 10−8 M glargine (Fig. 4B, top). By immunoprepitation with a polyclonal anti-IR antibody in HAEC, we could demonstrate a band corresponding to the IR β-subunit but no consistent phosphorylation by IGF-I or insulin (data not shown).
[3H]thymidine incorporation into DNA.
DNA synthesis was determined by [3H]thymidine incorporation in HMVEC (Fig. 5). The incorporation of [3H]thymidine was stimulated by addition of IGF-I and was highly significant at 10−7 M (P < 0.002). Neither insulin nor glargine had any significant effect.
The effect of IGF-I, glargine, and insulin on glucose metabolism in HMVEC cells was studied as [3H]glucose incorporation into the cells (Fig. 6). IGF-I significantly stimulated glucose incorporation at concentrations of 10−7 M (P = 0.008), 10−8 M (P = 0.02), and 10−9 M (P = 0.009), whereas insulin or glargine had no significant effect.
We show here that human micro- and macrovascular endothelial cells both have predominantly IGF-IR and less IR when measured by gene expression and ligand binding. To our knowledge, there are no previous studies addressing the comparison of IGF-IR and IR mRNA gene expression in human endothelial cells. There is only one report, stating that cultured human retinal endothelial cell from diabetic and nondiabetic patients express IGF-IR mRNA (39). The results from our study show that the specific binding of 125I-IGF-I was severalfold higher than the specific binding of 125I-insulin in both HAEC and HMVEC, indicating a higher number of IGF-IR. Both human insulin and the new insulin analog glargine interacted with the IGF-IR with over thousand- and hundredfold less potency than IGF-I itself. By use of ligand binding, a high number of IGF-IR was previously reported in human glomerular endothelial cells (29). In human umbilical vein endothelial cells, which are often used as a model for studies of human endothelium because they are easily available, specific binding of insulin (4, 5) and IGF-I (45) has been reported. The number of IGF-IR was estimated to be higher than the number of IR in this type of cells (45). We found that the specific binding of IGF-I in HMVEC and HAEC fitted a one-site binding model with an affinity corresponding to the binding of IGF-I to isolated IGF-IR (23). Binding of IGF-I to IGF-binding proteins (IGFBPs) produced by endothelial cells (14) could interfere with our results. Our finding of a one-site binding model for IGF-I and the fact that IGF-I was displaced by insulin, which cannot bind to IGFBPs (37), strongly argues against binding of IGF-I to IGFBPs. The specific binding of insulin in HMVEC and HAEC fitted a two-site binding model with affinities of the binding sites corresponding to a binding of insulin to the IR and IGF-IR, respectively (8, 23).
We found that the IGF-IR β-subunit was phosphorylated by both IGF-I and insulin in HAEC and HMVEC. In a previous study of HAEC treated with a high concentration of insulin (5.0 × 10−8 M) no insulin-induced IGF-IR phosphorylation was observed, whereas different concentrations of glucose and glucosamine induced a reduction in insulin-stimulated IR tyrosine phosphorylation (12). The results are difficult to interpret, because the presence of IR or IGF-IR protein in Western blot gel was not shown. IR protein and IR autophosphorylation by insulin at a low concentration in HAEC were reported by Aljada et al. (1). We were able to demonstrate the presence of IR in HAEC by Western blot but did not obtain convincing data regarding phosphorylation of IR β-subunit by insulin or IGF-I (data not shown). The insulin analog glargine at concentrations of 10−8 and 10−6 M was found to phosphorylate the IGF-IR β-subunit when tested in HMVEC, which shows that glargine activates the IGF-IR in these cells. That 10−8 M glargine, but not 10−8 M insulin, phosporylated the IGF-IR is in agreement with the higher affinity of glargine compared with insulin for the IGF-IR. This conclusion must be drawn with caution, because the results were not obtained in the same experiment. It has been shown previously as well as in this study that insulin can phosphorylate the IGF-IR (18). To our knowledge, the effect of glargine on IGF-IR phosphorylation has not been shown previously.
With regard to the IGF-IR, there are two properties that may create confusion when one is interpreting experimental results. First at high, supraphysiological concentrations, insulin binds to and activates the IGF-IR, as shown in this and other studies (3, 25). Many in vitro experiments of insulin action on vascular cells have been performed at very high insulin concentrations, not occurring in vivo, where stimulation of IGF-IR could be interpreted as insulin effects (2, 44). Here, we show that, in HMVEC, only IGF-I significantly stimulates DNA synthesis and glucose metabolism at low concentrations (13), indicating that the effects are elicited by IGF-IR and not by IR. Second, it has been demonstrated that, in tissues coexpressing insulin and IGF-IR, the two receptors form hybrid receptors composed of an IR αβ-heterodimer and an IGF-IR αβ-heterodimer, which functionally behave like IGF-IR (26, 30, 38). Even if gene expression of IR and IR protein can be demonstrated, there need not be any functioning IR. Our result showing a predominance of gene expression of IGF-IR and ligand binding mainly of IGF-I suggests that there was largely IGF-IR or possibly hybrid receptors in the human micro- and macrovascular endothelial cells.
What could be the consequence if the endothelial cells respond to either IR or IGF-IR? With regard to the prevalence of IGF-IR in human endothelial cells, it is of great interest that several conditions with low circulatory IGF-I are associated with vascular disease. IGF-I levels are low in type 1 diabetes mellitus, probably due to insufficient insulinization of the liver (11, 15). In normal aging, IGF-I levels decrease with increasing age throughout adulthood (24, 28). Patients with low IGF-I due to growth hormone deficiency are characterized by increased mortality attributable to cardiovascular disease (7, 33). In epidemiological studies, association between low IGF-I levels and increased prevalence of vascular diseases supports the hypothesis of IGF-I being involved in the pathogenesis of ischemic heart disease (16, 20).
Another aspect of our study is related to safety in using insulin analogs. We show here that glargine clearly behaves differently from human insulin with regard to its affinity for IGF-IR being ∼10-fold higher than that of human insulin. There is only one report in human osteosarcoma cells where glargine was 6.5 times more potent than human insulin in binding to the IGF-IR (23). Although in our study 10−8 M glargine had a weak effect on phosphorylation of the IGF-IR, there was no significant effect of glargine on DNA synthesis at a concentrations of 10−9 to 10−7 M. Even if glargine has a higher affinity than insulin for the IGF-IR, it will probably not have any mitogenic effect in vivo, because the concentrations of glargine reached in vivo are not high enough to considerably affect the IGF-IR (35). It cannot be excluded, however, that glargine could have some effect on hybrid receptors (30).
The result of this study clearly demonstrates that human micro- and macrovascular endothelial cells express IR and IGF-IR. As for the IR, we were able to demonstrate specific ligand binding and also IR protein but no significant effect of insulin. IGF-I activates the IGF-IR, and it can be concluded that human micro- and macrovascular cells possess functioning IGF-IR that can have a great impact in the pathogenesis of micro- and macroangiopathy.
Financial support was obtained from Landstinget Östergotland, the Swedish Medical Research Council (04952), the Swedish Diabetes Association, and Barndiabetes Fonden.
We are grateful to Anna-Kristina Granath and Margareta Karlsson for excellent technical assistance.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 by American Physiological Society