Impaired Generation of Reactive Oxygen Species in Leprechaunism Through Downregulation of Nox4
Abstract
Leprechaunism features a clinical constellation characterized by extreme insulin resistance, growth retardation, and several distinct developmental abnormalities. One puzzling observation about leprechaunism is that mutations in the insulin receptor gene frequently associated with this syndrome cannot account for the aberrant responses of cultured cells to other growth factors. Here we report that the generation of reactive oxygen species (ROS) is impaired in cells from leprechaunism patients, thus shedding new light on this issue. Stimulation of patients’ skin fibroblast cells with platelet-derived growth factor (PDGF) resulted in a lower-level tyrosine phosphorylation of cytosolic proteins compared with that seen in normal cells. In addition, consistent with the hypothesis that ROS mediate the level of tyrosine phosphorylation of cytosolic proteins through inactivation of protein tyrosine phosphatases (PTPases), patient fibroblast cells showed a significantly higher phosphatase activity than normal cells. We further showed that the lower-level tyrosine phosphorylation in response to growth factors results from the downregulation of an NADPH oxidase, Nox4, which in turn results in the reduction of ROS generation. Ectopic expression of Nox4 in the patient fibroblast cells consistently restored PDGF-induced ROS production and regulation of PTPase activities. Taken together, these data provide insight into the mechanisms through which growth retardation is associated with leprechaunism syndrome.
- DCF-DA, 2′,7′ dichlorofluorescein diacetate
- DMEM, Dulbecco’s modified Eagle’s medium
- DTT, dithiothreitol
- EGF, epidermal growth factor
- NOXO1, Nox organizing protein 1
- PDGF, platelet-derived growth factor
- PTPase, protein tyrosine phosphatase
- RMS, Rabson-Mendenhall syndrome
- ROS, reactive oxygen species
Leprechaunism is a rare congenital syndrome characterized by dysmorphic features (elfin-like face, low-set ears, prominent eyes, thick lips), decreased subcutaneous fat, hirsutism, acanthosis nigricans, and intrauterine and neonatal growth retardation (1–6). The metabolic abnormalities include severe insulin resistance, with hyperinsulinemia associated with postprandial glucose intolerance. Several cases of mutations in the insulin receptor gene have been reported among leprechaunism patients. These patients show a severe growth restriction that can be attributed to the lack of intrinsic activity of the insulin receptor (7). However, fibroblasts from patients with leprechaunism fail to respond to not only insulin but also other growth factors; specifically, IGF-I, platelet-derived growth factor (PDGF), and epidermal growth factor (EGF) fail to elicit normal responses in patient cells (8–11). Also, cells from these patients grow more slowly than normal cells in culture, reflecting an impaired response to growth stimulation (4,8). It is unclear how mutations in the insulin receptor gene are related to the aberrant responses of cultured fibroblasts to growth factors (8); it is possible that such defects contribute to the growth restriction observed in this syndrome.
Reactive oxygen species (ROS) include a wide variety of molecules, including singlet oxygen, the superoxide anion radical (O.̄2), H2O2, lipid peroxides, the thiylperoxyl radical, the ferryl radical, and the hydroxyl radical (OH·). ROS are generally considered cytotoxic because they can cause oxidative damage to cellular components (12). However, the generation of ROS also appears to contribute to receptor-mediated cell signaling through the regulation of protein phosphorylation (12,13). Exogenous addition of a catalase abolishes PDGF- and EGF-mediated tyrosine phosphorylation in smooth muscle cells and A431 cells, respectively (14,15). Recent studies have consistently demonstrated that ligand-induced H2O2 can specifically and reversibly regulate the activity of protein tyrosine phosphatases (PTPases) (12,16,17). Under an equilibrium state between opposing actions of protein kinases and phosphatases in cells, the inhibition of phosphatases by H2O2 shifts the equilibrium toward phosphorylation.
Receptor-mediated ROS production has been studied extensively in phagocytic cells. The enzyme NADPH oxidase in these cells is composed of at least five protein components, including two transmembrane flavocytochrome b components (gp91phox and p22phox) and four cytosolic components (p47phox, p67phox, and p40phox, and rac protein) (18,19). Exposing resting phagocytic cells to an appropriate stimulus results in extensive phosphorylation of the cytosolic components of NADPH oxidase and their association with the transmembrane flavocytochrome b components (20–23). The assembled oxidase complex catalyzes the transfer of an electron to molecular oxygen to yield the superoxide anion, which is then spontaneously or enzymatically converted to H2O2. The homolog of gp91phox in nonphagocytic cells, cloned from a colon cDNA library, is Nox1 (24). Additional homologs have been subsequently cloned in various cell types, including nonphagocytic cells, clearly suggesting a role of ROS in cellular processes other than phagocytosis. The family, called the Nox gene family, currently consists of seven human homologs: Nox1, gp91phox (Nox2), Nox3, Nox4, Nox5, Duox1, and Duox2 (25–28). Nox1, Nox3, and Nox4 are, like gp91phox, ∼65 kDa in size. Duox and Nox5 enzymes contain gp91phox homology domain as well as calcium-binding motifs (25–28). Although a typical mammalian cell contains several homologs of gp91phox, how they function as part of cell signaling pathway awaits further studies.
Several reports have indicated an association between insulin signaling and ROS. The elimination of ROS by the addition of antioxidants induces insulin resistance, and exogenous addition of hydrogen peroxide mimics the stimulatory effects of insulin on glucose transport and lipid synthesis (29,30). However, to date no study has associated the importance of the reduced ROS generation with the pathogenesis of insulin resistance such as leprechaunism. Moreover, mutations in the insulin receptor gene associated with this syndrome cannot account for the aberrant responses of cultured cells to other growth factors. Here we present a novel observation linking ROS generation to insulin resistance and providing a possible mechanism contributing to the unexplained growth retardation among patients with leprechaunism.
RESEARCH DESIGN AND METHODS
Cell culture and immunoblot analysis.
Primary skin fibroblasts obtained from skin biopsies of normal and leprechaunism patients were maintained in high-glucose Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 20% fetal bovine serum and 1% antibiotics at 37°C in a 5% CO2 solution. Skin fibroblast cells were plated in six-well plates at 105 cells/well. Cells were deprived of serum for 16 h and then incubated for 10 min at 37°C in the absence or presence of PDGF (50 ng/ml) or EGF (100 ng/ml). Incubation with PDGF or EGF was terminated by adding a lysis buffer (50 mmol/l Tris-HCl [pH 7.5], 1 mmol/l EDTA, 1% NP-40, 150 mmol/l NaCl, 1 μg/ml aprotinin, 2 μg/ml leupeptin, and phosphatase inhibitor cocktail II; Sigma). The proteins in lysates were separated by SDS-PAGE and visualized with immunoblot.
Adenoviral transduction of fibroblasts.
Control and recombinant adenoviruses encoding Nox4 construct were incubated with skin fibroblasts for 100 min at room temperature in DMEM containing 0.5% BSA and 0.5 μg polylysine/ml. After the incubation, adenovirus infection of the skin fibroblasts was performed by overnight incubation in DMEM containing 0.5% BSA. The next day, the medium was replaced with complete medium containing 20% fetal bovine serum. The cells were used for experiments 60 h posttransduction after they were starved for 12 h in serum-free medium (31).
Assay of intracellular H2O2 production.
Intracellular production of H2O2 was assayed after cells were stimulated with PDGF (50 ng/ml) or EGF (100 ng/ml; Upstate Biotechnology) for 10 min in serum-free DMEM. Dishes of confluent cells were washed with Hanks’ balanced salt solution and incubated for 5 min in the dark at 37°C with the same solution containing 5 μmol/l 2′,7′ dichlorofluorescein diacetate (DCF-DA; Molecular Probes). Analysis of intracellular H2O2 production with scanning confocal microscope (LSM 510; Carl Zeiss) was performed as previously described. All experiments were repeated at least five times (32).
PTPase assay.
Intracellular PTPase activity was assayed after cells were stimulated with PDGF (50 ng/ml), EGF (100 ng/ml), or insulin (100 nmol/l) for 10 min in serum-free DMEM. After being stimulated, the cells were frozen in liquid nitrogen and moved to an anaerobic chamber. The cells were scraped in lysis buffer (1% Triton X-100, 0.5% NP-40, 150 mmol/l NaCl, 10% glycerol, 1 mmol/l EDTA, and 2 mmol/l EGTA in 20 mmol/l HEPES buffer [pH 7.0]) containing 10 mmol/l iodoacetate. The cell lysates were incubated for 30 min in the dark to achieve complete alkylation of free thiols. The labeling was quenched by adding 20 mmol/l dithiothreitol (DTT). Then the lysates were centrifuged at 14,000 rpm, and the concentration of protein in lysate was assayed using a bicinchoninic acid assay. The PTPase activities were measured in a 96-well plate coated with poly-(Glu4-pTyr) peptides using the Universal Tyrosine Phosphatase assay kit (MK-411; Takara Bio). PTPase activity was quantified from a standard curve obtained using the recombinant CD45 tyrosine phosphatase. PTPase activity, measured from three independent experiments, is expressed as relative PTPase activity per milligram of protein used (33).
Quantitative real-time PCR.
Quantification of human Nox1, Nox2, Nox4, Nox organizing protein 1 (NoxO1), and the glyceraldehyde-3-phosphate dehydrogenase expression level was performed by amplification of cDNA with the Applied Biosystems Model 7000 Sequence Detection System (PerkinElmer). Quantitative real-time PCR was performed using the TaqMan universal PCR master mix. Predesigned, gene-specific TaqMan probe and Primers were purchased from Applied Biosystems. The temperature profile for the reaction was 50°C for 2 min, 95°C for 10 min, and then 95°C for 15 s and 60°C for 1 min for 45 cycles. Copy numbers were calculated by the instrument software from standard curves generated from human Nox1, Nox2, Nox4, NoxO1, and glyceraldehyde-3-phosphate dehydrogenase templates.
RESULTS
Reduced tyrosine phosphorylation in the skin fibroblast cells from patients with insulin resistance in response to PDGF, EGF, or insulin.
To compare the phosphorylation of intracellular protein by PDGF or EGF, skin fibroblasts isolated from three patients with insulin resistance and from a healthy individual were stimulated by PDGF-BB or EGF for various lengths of times (5–60 min). Two of the fibroblast cell lines (NZ and Mt. Sinai) have mutations in the insulin receptor that abolish insulin binding and insulin-mediated cell signaling (34). No mutation was found in the exons of the insulin receptor gene in the third patient cell line (Rabson-Mendenhall syndrome [RMS]), even though the typical dysmorphic features of leprechaunism and metabolic abnormalities associated with severe insulin resistance were present. The cells from the third patient had decreased insulin binding despite normal IGF-I binding (data not shown). Fibroblast cells from the RMS patient showed a lower overall tyrosine phosphorylation level of intracellular proteins after PDGF and EGF stimulation compared with that of normal cells (Figs. 1A and B). We observed that the other two fibroblast cell lines (NZ and Mt. Sinai) showed similarly lowered level tyrosine phosphorylation of intracellular proteins after PDGF stimulation (data not shown). PDGF induces phosphorylation of several tyrosine residues in the cytosolic region of the PDGF receptor. Phosphorylated Y1021 serves as a binding site for the SH2 domain of PLC-γ1, which is then phosphorylated at tyrosine residues 771, 783, and 1,254 by the PDGF receptor tyrosine kinase, leading to activated lipase activity (35). The stimulation of patient cells with PDGF resulted in a clearly reduced phosphorylation of PDGF receptor and PLC-γ1 (Fig. 1C).
Reduced tyrosine phosphorylation by PDGF or EGF in the skin fibroblast cells from leprechaunism patients. A and B: Skin fibroblast cells from a normal individual and an RMS patient were plated in six-well dishes. Cells were deprived of serum for 16 h and incubated at 37°C in the absence or presence of PDGF (50 ng/ml) or EGF (100 ng/ml). Cell lysates were subjected to immunoblot analysis with anti-phosphotyrosine (4G10; Upstate Biotechnology), The filter was reprobed with antibodies to actin. C: Cell lysates were prepared after being incubated with PDGF for 10 min and immunoblotted with antibodies to PDGFR-PY1021 (LabFrontiers, Seoul, South Korea). The cell lysates were immunoprecipitated with anti−PLC-γ1 antibody, and the resulting precipitates (IP) were subjected to immunoblot analysis with anti-phosphotyrosine antibody (middle panel). Lysates were also directly subjected to immunoblot analysis with antibodies to actin (lower panel). WB, Western blot.
Impaired ROS generation by growth factors in insulin resistance.
Previously, we reported that elimination of ROS by the addition of a catalase abolished tyrosine phosphorylation of cytosolic proteins, including PLC-γ1 (14,15). Therefore, we questioned whether the low level of tyrosine phosphorylation after PDGF or EGF stimulation in patients’ cells was correlated with reduced ROS generation. To test our hypothesis, we measured the ROS induction by using DCF-DA oxidation with a confocal microscope (15,32). ROS generation was induced in normal fibroblasts in response to PDGF, EGF, or insulin, whereas cell lines from the three patients (RMS, NZ, and Mt. Sinai) failed to generate ROS after being stimulated with these agonists (Fig. 2A).
A: Impaired ROS generation by growth factor in leprechaunism patient cells. Intracellular production of H2O2 was assayed after RMS cells were stimulated with PDGF (50 ng/ml), EGF (100 ng/ml), or insulin (100 nmol/l) in serum-free DMEM. The mean fluorescence intensity of DCF in experiments was measured and expressed in relation to that of unstimulated cells. Data are means ± SE from five independent experiments. B: Cells were serum starved for 16 h and stimulated with PDGF (50 ng/ml), EGF (100 ng/ml), or insulin (100 nmol/l) for 10 min. After being stimulated, the cells were frozen in liquid nitrogen and moved to an anaerobic chamber. The cells were scraped in lysis buffer (1% Triton X-100, 0.5% NP-40, 150 mmol/l NaCl, 10% glycerol, 1 mmol/l EDTA, and 2 mmol/l EGTA in 20 mmol/l HEPES buffer [pH 7.0]) containing 10 mmol/l iodoacetate. The cell lysates were subjected to a PTPase assay as described in research design and methods. Data are means ± SE from three independent experiments. *P < 0.01 vs. control; **P < 0.001 vs. control.
Several recent reports have suggested that ROS act as intracellular messengers modulating the extent of protein phosphorylation through reversible inactivation of PTPases (12,13,16,36). Cysteine residues in the active center of PTPase are easily oxidized by ROS, leading to the inhibition of phosphatase activity. Stimulation of cells with EGF and PDGF results in the inhibition of PTP-1B and SHP-2 activities, respectively (16,36). We first measured PTPase activity in the absence or presence of PDGF, EGF, or insulin. The free sulfhydryl group of cysteine in PTPase is conjugated by iodoacetamide, whereas the sulfenic group, the oxidized form of the sulfhydryl group by intracellular generated H2O2, is not reactive with this modifying reagent (iodoacetamide). The oxidized cysteine residue of PTPase is subsequently reduced by adding reducing agents such as DTT, which then restores PTPase activity. The activities of oxidized PTPase (after recovery from DTT reduction) in total cell lysates from normal or RMS fibroblasts were measured using poly-(Glu4-pTyr) peptides as the substrate. In the normal fibroblasts, oxidation of PTPase by PDGF, EGF, and insulin was increased by 1.48-, 1.44-, and 1.2-fold, respectively, compared with unstimulated cells (Fig. 2B). However, the oxidation of PTPase by three agonists in RMS fibroblasts showed no increase (Fig. 2B). These data suggest that the lower level of tyrosine phosphorylation by growth factors resulted from the failure of PTPase activity regulation in the three patients’ fibroblasts.
Downregulation of Nox4 is responsible for reduced ROS generation in insulin resistance.
It is likely that the three different agonists (PDGF, EGF, and insulin) failed to increase ROS generation in cells from three independent patients with insulin-resistant cells because of a single common mechanism. In other words, the results suggest that a common signaling component for ROS generation may be deficient in the patients’ cells. The enzyme NADPH oxidase consists of multiple protein components (gp91phox, p22phox, p47phox, p67phox, p40phox, and rac) (18–23). We investigated the expression level of NADPH oxidase components and accessory proteins in the patients’ cells. The expression of most components of receptor-mediated ROS generation was comparable with that of normal cells, but the expression levels of Nox1, Nox4, and NoxO1 were decreased in the patients’ cells (Figs. 3 and Table 1). Quantitative real-time PCR indicated that Nox4 is the predominant form and Nox1 is a minor one in skin fibroblasts. The expression of Nox4 and Nox1 in cell lines from three independent patients was decreased by 65 and 50%, respectively, compared with that in normal cells. Expression levels of Nox2 and Nox5 were unchanged, and Nox3 was not detected in either normal or insulin-resistant cells. Moreover, the expression of NoxO1 was decreased by 60% in patient cells, whereas NADPH activator 1 was not detected in either of the patient or normal cells (Table 1).
Downregulation of Nox4 is responsible for the reduced ROS generation in leprechaunism. The cell lysates of the each patient sample were analyzed by immunoblot with an anti-p67phox, -p47phox, -p22phox, and -Rac antibody. The filter was reprobed with antibodies to actin. WB, Western blot.
Expression levels of various NADPH oxidase components
We next investigated whether the production of ROS in the RMS patient’s cells could be restored by ectopic expression of Nox4. Infection of the patient’s fibroblasts with adenovirus encoding Nox4 resulted in an increased Nox4 mRNA level of >3.5-fold compared with cells infected with control adenovirus (data not shown). The overexpression of Nox4 resulted in increased basal ROS generation and, more importantly, the ROS level in response to PDGF or EGF stimulation in the RMS fibroblasts (Fig. 4). We showed that low tyrosine phosphorylation in the patient’s cells was due to the failure of PTPase regulation resulting from downregulation of Nox (Fig. 2B). We tested whether the recovery of ROS generation through the introduction of Nox4 to RMS fibroblasts could restore PTPase activity. In normal fibroblasts, oxidation of PTPase was increased by 1.5-fold after PDGF stimulation compared with unstimulated cells (Fig. 5A). Oxidation of PTPase by PDGF was increased by twofold in RMS fibroblasts overexpressing Nox4 compared with RMS parental cells (Fig. 5A). These results indicated that the other components necessary for ROS generation except Nox4 were intact in insulin-resistant cells and that ectopic expression of Nox4 is the key element in the recovery of ROS generation and PTPase activity in response to growth factor stimulation.
A: Normal and RMS skin fibroblast cells were infected with adenovirus encoding β-galactosidase (β-gal) as a control or recombinant Nox4 construct (Ad-Nox4). The cells were then deprived of serum for 16 h, incubated for 10 min at 37°C in the absence or presence of PDGF (50 ng/ml) or EGF (100 ng/ml), and assayed for the production of H2O2. B: The mean fluorescence intensity of DCF in experiments was measured and expressed in a value relative to that of unstimulated cells. Data are means ± SE from three independent experiments.
A: Normal and RMS skin fibroblast cells were infected with adenovirus encoding β-galactosidase (β-gal) as a control or recombinant Nox4 (Ad-Nox4). Cells were serum starved for 16 h before being stimulated with PDGF (50 ng/ml) for 10 min. Total cell lysates were prepared and PTPase activity was assayed as described in research design and methods. Data are means ± SE from three independent experiments. *P < 0.01; **P < 0.001. B and C: Both cells were then deprived of serum for 16 h and incubated for 10 min at 37°C in the absence or presence of PDGF (50 ng/ml; B) or EGF (100 ng/ml; C). Cell lysates were subjected to immunoblot analysis with anti-phosphotyrosine antibody (4G10), and the filter was reprobed with antibodies to actin. WB, Western blot.
To test whether the overexpression of Nox4 could restore tyrosine phosphorylation after growth factor stimulation, we performed immunoblot analysis with an antibody against phosphotyrosine (4G10, UBI). As shown in Figs. 5B and C, the tyrosine phosphorylation by PDGF or EGF increased in the RMS fibroblast overexpressing Nox4 compared with the RMS fibroblast infected with control β-galactosidase. These results suggested that the recovery of ROS generation by the overexpression of Nox4 can restore tyrosine phosphorylation after PDGF and EGF stimulation in patient fibroblast cells.
DISCUSSION
Several lines of evidence implicate H2O2 as an intracellular messenger modulating the extent of protein phosphorylation (12,16,17,36). Exogenous H2O2 mimics the effect of a growth factor on the induction of PTPase and mitogen-activitated protein kinase activation (35). The link between ligand-induced H2O2 generation and protein phosphorylation was further strengthened by the observation that scavenging of H2O2 by exogenously introduced catalase completely inhibited PDGF- and EGF-induced tyrosine phosphorylation in rat vascular smooth muscle cells and A431 cells, respectively (14,15). Moreover, recent studies have demonstrated that ligand-induced H2O2 can specifically and reversibly regulate the activity of PTPases (12,13,16,17). Under an equilibrium state maintained by opposing actions of protein kinases and phosphatases in cells, the inhibition of phosphatases by H2O2 shifts the equilibrium toward phosphorylation. We have shown here that the low levels of tyrosine phosphorylation by growth factor in cells from patients with insulin resistance result from impaired ROS generation (Fig. 2A). Specifically, lowered inhibition of phosphatase activity resulting from impaired ROS generation led to a lowered tyrosine phosphorylation of cytosolic proteins in response to PDGF (Figs. 1 and 2).
NADPH oxidase plays a key role in ROS generation in response to a number of growth factors (19,32). The catalytic component of NADPH oxidase is comprised of several proteins encoded by the Nox gene family. Several lines of evidence suggest that Nox regulates cellular responses to various growth factors (19,28,32). Mutational analysis of the PDGF receptor has revealed that phosphatidylinositol 3-kinase activation is essential for growth factor−induced ROS production (15). We previously reported a sequential mechanism by which growth factor stimulation induces the production of ROS. The binding of growth factors to their receptors results in the activation of phosphatidylinositol 3-kinase, the products of which (PtdIns[3,4,5]P3 and PtdIns[3,4]P2) then bind to the pleckstrin homology domain of Nox-associated β-Pix and stimulate the GDP-GTP exchange activity of β-Pix. The activated β-Pix converts Rac1-GDP to Rac1-GTP, which then associates with Nox1 to promote the electron transfer from NADPH to molecular oxygen (32). Nox1 and Nox4 are also involved in angiotensin II−induced signaling in smooth muscle cells and insulin-dependent signaling in adipose tissues (31,37). Overexpression of Nox and increased ROS levels are seen in cells and tissues with a strong mitogenic activity. Nox overexpression has been reported in human cancer cells, including colon cancer (Nox1) (24), melanomas (Nox4) (38), and prostate cancer (Nox4 and Nox5) (38,39).
Nox1 activity is regulated by accessory proteins such as NoxO1 and NADPH activator 1 (40,41), whereas Nox4 is constitutively active and does not require accessory proteins (42,43). It has been shown that the regulation of Nox isozymes as well as their expression level plays an important role in cellular proliferation (19,42). The level of Nox4-induced ROS generation is likely determined mainly by the expression level of Nox4 itself. Therefore, downregulation of Nox4 in patient cells likely contributes to the impaired ROS generation in response to growth factor stimulation, leading to complicated growth retardation (Table 1 and Fig. 3). The introduction of Nox4 in the RMS cells consistently restored ROS production and its response to growth factor stimulation such as tyrosine phosphorylation of cytosolic proteins (Figs. 4 and 5).
Leprechaunism and RMS are characterized by different degrees of growth restriction (4,6,8). Animal studies support the idea that the insulin receptor has a direct action on growth, although the mechanism by which this occurs remains unclear (7). Our findings indicate that cells from patients with insulin resistance due to abnormal insulin receptor signaling have impaired ROS production. This might contribute to the growth retardation of patients with leprechaunism or RMS. Abnormal ROS production might also explain the abnormal cross-talk between different tyrosine kinase receptors observed in cells from these patients, and such aberrant signaling events might contribute to some aspects of these complex syndromes.
Acknowledgments
This work was supported by the Korea Science and Engineering Foundation (KOSEF) through the Center for Cell Signaling Research at Ewha Womans University, 21C Frontier Functional Proteomics Project (FPR05C2-510) from the Korea Ministry of Science and Technology, a grant from Samsung Biomedical Research Institute (SBRI C-A3-112-2 to D.K.J.), and the Ewha-SK Fund (to Y.S.B.). H.S.P. is a recipient of a BK21 scholarship and of KRF-2005-216-C100126 from the Korea Research Foundation.
We thank Dr. Goldstein for adenovirus encoding Nox4 and Drs. Sue Goo Rhee and Jaesang Kim for their discussion and critical reading of the manuscript.
- Received September 20, 2004.
- Accepted August 8, 2005.
- DIABETES
REFERENCES
In this Issue
