Oxidative stress and adipogenesis play key roles in the pathogenesis of Graves’ orbitopathy (GO). In this study, the therapeutic effects of caffeine on the reduction of oxidative stress and adipogenesis were evaluated in primary cultured GO orbital fibroblasts in vitro. Orbital fibroblasts were cultured from orbital connective tissues obtained from individuals with GO. Intracellular reactive oxygen species (ROS) levels induced by hydrogen peroxide or cigarette smoke extract and the expression of anti-oxidative enzymes were measured after caffeine treatment. After adipogenic differentiation and caffeine treatment, cells were stained with Oil Red O and the levels of peroxisome proliferator activator γ (PPARγ), C/EBPα, and C/EBPβ were determined by western blot analysis. Hydrogen peroxide and cigarette smoke extract increased the levels of intracellular ROS and anti-oxidative enzymes, which decreased in a dose-dependent manner upon pretreatment with caffeine in GO orbital fibroblasts. Oil Red-O staining results revealed a decrease in lipid droplets; furthermore, PPARγ, C/EBPα, and C/EBPβ protein expression levels were inhibited upon treatment with caffeine during adipocyte differentiation. In conclusion, caffeine decreased oxidative stress and adipogenesis in GO orbital fibroblasts in vitro. These findings may contribute to the development of new types of caffeine-containing pharmacological agents for use in the management of GO.

GRAVES’ ORBITOPATHY (GO) is an autoimmune disorder of the orbit involving the infiltration of T cells, B cells, plasma cells, and macrophages. The pathogenesis of GO involves inflammation of orbital connective tissue that leads to the enlargement of extraocular muscles and orbital adipose tissue. Such tissue expansion within the bony orbits is thought to be responsible for the progression and complications of GO.

At the center of this pathogenesis lies inflammation and oxidative stress. Infiltration of inflammatory cells to the orbit triggers and propagates tissue expansion. Oxygen free radicals trigger fibroblast proliferation [1] and the abundance of thyroid hormones in Graves’ disease renders orbital fibroblasts more vulnerable to oxidative damage [2]. Clinically, cigarette smoking is the most important environmental risk factor for the development and progression of GO [3, 4], suggesting that oxidative stress is one of the key effectors in its pathogenesis.

Caffeine (1,3,7-trimethylxanthine) is a plant alkaloid found in coffee, tea, chocolate, cola, and other soft drinks commonly consumed around the world [5]. It is absorbed rapidly and completely from the gastrointestinal tract and is detectable in all body fluids including in umbilical cord blood [6, 7]. Numerous studies have determined that it acts as a free radical scavenger, taking part in antioxidant activities in vitro and in vivo in both animals and humans [8-10]. Furthermore, its anti-adipogenic properties in various cell types in vitro have shown that caffeine inhibited the expression of adipogenesis-related cytokines and transcription factors, resulting in a decrease in adipocyte differentiation and intracellular lipid accumulation [5, 11]. Given the fundamental role of oxidative stress and adipogenesis in the pathogenesis of GO, we investigated the therapeutic effect of caffeine in an in vitro model of GO.

Dulbecco’s modified Eagle’s medium (DMEM), fetal bovine serum (FBS), penicillin, and gentamicin were purchased from Hyclone Laboratories, Inc. (Logan, UT, USA). Caffeine was purchased from Sigma-Aldrich Corp. (St. Louis, MO, USA). The 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt (MTS) assay solution was purchased from Promega Corporation (Madison, WI, USA). Cigarette smoke extract (CSE) was freshly prepared within an hour of each experiment from commercially available filtered cigarettes [Marlboro 20 class A cigarettes (8.0 mg tar; 0.7 mg nicotine); Philip Morris Korea, Inc., Seoul, Korea] as described in our previous study [12]. Recombinant human IL-1β was purchased from R&D Systems (Minneapolis, MN, USA) and Oil Red O was purchased from Sigma-Aldrich Corp.

Orbital adipose/connective tissue explants were obtained as surgical waste during decompression surgery in five patients with GO (three females, two males; 29–51 years of age). Normal control tissues were harvested during upper-lid blepharoplasty from the post-septal area of five individuals with no history or clinical evidence of thyroid disease or GO (three females, two males; 31–58 years of age). All five patients with GO achieved stable euthyroidism at the time of surgery, at which point their clinical activity scores were less than three. Furthermore, none of the patients with GO received steroid treatment or radiotherapy for at least three months before surgery. The Institutional Review Board of Severance Hospital, Yonsei University College of Medicine (Seoul, Korea) approved the study and written informed consent was obtained from all participants after explanation of the nature and possible consequences of the study. This study followed the tenets of the Declaration of Helsinki.

Primary cultures of orbital fibroblasts were established as described in our previous study [13]. Briefly, minced tissue was placed directly in 1:1 DMEM:F12 medium with 20% FBS and antibiotics. When the growth of fibroblasts was observed, monolayers were passaged serially with trypsin/ethylenediaminetetraacetic acid solution and cultures were maintained in DMEM with 10% FBS and antibiotics. The strains were stored in liquid nitrogen until further analysis; cells between the second and fifth passages were used.

The anti-adipogenic effect of caffeine was evaluated using our previously reported adipocyte differentiation protocol for GO orbital fibroblasts involving dexamethasone, rosiglitazone (10 μM; Cayman Chemical, Ann Arbor, MI, USA), and IL-1β (10 ng/mL) treatment [13, 14]. Briefly, after 10 days of adipogenic differentiation co-treated with serial concentrations of caffeine (0.1–5 mM), differentiated adipocytes were visualized with Oil Red O staining and expression of adipogenic transcription factors were evaluated with western blotting.

To evaluate the effect of caffeine on cell viability of orbital fibroblasts, primary cultured orbital fibroblasts obtained from patients with GO were seeded on 24-well culture plates (1 × 10⁴ cells/well) and treated with various concentrations of caffeine (0.01 mM, 0.1 mM, 0.5 mM, 1 μM, 2 mM, and 5 mM) for 48 h and 72 h. Thereafter, MTS solution was added and the plate was incubated again for 4 h under the same conditions. The absorbance of the dye was measured at 490 nm using an ELISA plate reader. Cell viability is expressed as a percentage relative to untreated control cells.

ROS release was determined with 5-(and 6)-carboxy-20,70-dichlorodihydrofluorescein diacetate (H2DCFDA; Invitrogen/Thermo Fisher Scientific, Carlsbad, CA, USA), an oxidant-sensitive fluorescent probe, as we described previously [12]. Briefly, primary cultured orbital fibroblasts were pretreated with various concentrations of caffeine for 24 h. The culture medium was then removed, and the cells were washed with PBS, incubated with 10 mM H2DCFDA at 37°C for 30 min, then stimulated with CSE (2%) or H₂O₂ (10 μM) for 30 min [12, 15, 16]. The fluorescently stained cells were examined microscopically at 40× and were quantified via flow-cytometric analysis.

Western blot analysis was performed as described previously [17]. To assess the induction of oxidative stress by CSE and H₂O₂ and its inhibition by caffeine, confluent orbital fibroblasts were pre-exposed to serial concentrations of caffeine for 24 h, followed by stimulation with CSE or H₂O₂ for 24 h. Expression levels of manganese superoxide dismutase (Mn-SOD), copper-zinc superoxide dismutase (Cu/Zn-SOD), thioredoxin (Trx), and heme oxygenase-1 (HO-1) were assessed to evaluate oxidative stress levels. To evaluate the expression of adipogenic transcription factors after adipogenic differentiation, the expression levels of peroxisome proliferator activator gamma (PPARγ), C/EBPα, and C/EBPβ were evaluated in GO cells after ten days of adipocyte differentiation. Activation of intracellular signalling proteins was assayed via western blot analysis of phosphorylated and total AKT, extracellular signal-regulated kinase (ERK), nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB), c-Jun NH(2)-terminal kinase (JNK), and p38 protein. The relative amount of protein in each immunoreactive band was quantified by densitometry and normalised to the concentration of β-actin in the same sample.

Anti-Mn-SOD, anti-Zn/Cu-SOD, anti-Trx, anti-HO-1, anti-peroxisome proliferator activator gamma (PPARγ), anti-C/EBPα, anti-C/EBPβ, anti-NF-κB, p38, and anti–β-actin antibodies were all obtained from Santa Cruz Biotechnology (Dallas, TX, USA) and antibodies against AKT, ERK, and JNK were produced by Cell Signaling Technology (Danvers, MA, USA).

After ten days of adipocyte differentiation of orbital fibroblasts, cells were stained with an Oil Red O working solution as described previously [13, 14]. Stained cells were visualised and photographed at 40×, then quantified using a spectrophotometer at 490 nm according to the manufacturer’s instructions. Experiments for the quantitative assessment of adipogenic differentiation were performed in duplicate using cells from different donors; the results were normalised to the absorbance of untreated differentiated control cells.

All experiments were performed using cells from at least three different samples and the samples were assayed in duplicate each time. Differences in parameter estimates between the experimental and control groups were assessed by the Student’s t-test or Wilcoxon rank-sum test using R version 3.1.2 (R Foundation, Vienna, Austria). Values of p < 0.05 were considered significant.

To determine non-toxic concentrations of caffeine in orbital fibroblasts, an MTS assay was performed. Orbital fibroblasts from patients with and without GO were treated with caffeine at concentrations ≤5 mM for 48 h and 72 h. The 0.01 to 5 mM range of caffeine did not decrease cell viability below 95% in both normal and GO orbital fibroblasts in the 48 h treatment (Fig. 1, data for non-GO orbital fibroblasts are not shown).

Fig. 1

Effect of caffeine on the viability of orbital fibroblasts

Orbital fibroblasts from individuals with GO were seeded in 24-well culture plates and treated with various concentrations (0–5 mM) of caffeine for 48 h (A) and 72 h (B). After treatment, MTS assays were used to evaluate viability. Assays were performed in duplicate with cells from three different donors. Results are expressed as percentages of untreated control values and are presented as means ± standard deviation.

To demonstrate the anti-oxidative effect of caffeine in GO orbital fibroblasts, oxidative stress was induced by 10 μM H₂O₂ or 2% CSE with or without 0.01–5 mM caffeine pretreatment for 24 h. Treatment with 10 μM H₂O₂ or 2% CSE for 30 min significantly increased ROS production relative to that in untreated control cells, which was reduced by caffeine pretreatment in a dose-dependent manner (Fig. 2A). Similarly, upon fluorescent staining of cellular ROS, a significant reduction in fluorescently stained cells was observed after treatment with caffeine (40×; Fig. 2B).

Fig. 2

Effect of caffeine on H₂O₂- and CSE-induced generation of intracellular ROS in orbital fibroblasts from GO patients

Orbital fibroblasts from GO patients (n = 3) were treated with 10 μM H₂O₂ or 2% CSE for 30 min with or without pretreatment with caffeine for 24 h. ROS levels were measured via flow cytometry using H₂DCFDA (A) and microscopically visualised at 40× (B). The results are expressed as percentages of the untreated control values and presented as means ± standard deviation. Assays were performed in duplicate with cells from three different GO samples; data from a representative experiment are shown (*p < 0.05 and **p < 0.01 versus H₂O₂- or CSE-treated cells without caffeine pretreatment).

The relative expression levels of Mn-SOD, Cu/Zn-SOD, and HO-1 increased upon treatment with 10 μM H₂O₂ or 2% CSE in GO orbital fibroblasts for 24 h, while the expression of Trx increased only upon treatment with CSE, not H₂O₂. Upon pretreatment with a serial concentration of caffeine, we found that caffeine concentrations >1 mM decreased the expression of Mn-SOD, Cu/Zn-SOD, Trx, and HO-1 (Fig. 3).

Fig. 3

Effect of caffeine on H₂O₂- and CSE-induced expression of SOD, Trx, and HO-1 in orbital fibroblasts from GO patients

Orbital fibroblasts from GO patients (n = 3) were treated with 10 μM H₂O₂ (A) or 2% CSE (B) for 24 h with or without pretreatment with 0.1–5 mM caffeine for 24 h. The cell lysates were subjected to western blot analysis. Experiments were performed three times using different strains and samples were assayed in duplicate. Results are expressed as the relative density of each protein normalised to the level of β-actin and presented as means ± standard deviation. GO cells treated with neither CSE nor H₂O₂ were used as controls (*p < 0.05 and **p < 0.01 versus H₂O₂- or CSE- treated cells without pretreatment).

To determine the effect of caffeine on adipogenesis in orbital fibroblasts, GO cells were treated with caffeine during adipocyte differentiation. When caffeine (0.5–5 mM) was added to the adipogenic medium, it reduced the number of adipocytes and suppressed the accumulation of lipid droplets. The optical density of Oil Red O-stained cell lysates showed that caffeine-treated cells exhibited decreased absorbance at 490 nm (Fig. 4A). Western blot analysis was then performed to investigate whether caffeine affects the production of the adipogenic transcription factors PPARγ, C/EBPα, and C/EBPβ during adipogenesis. Caffeine had dose-dependent inhibitory effects on the expression of adipogenic transcription factors (Fig. 4B). Each experiment was performed in three GO cells from different patient samples and samples were assayed in duplicate.

Fig. 4

Effect of caffeine on adipogenesis in GO orbital fibroblasts

(A) Ten days of adipocyte differentiation was induced in primary cultured orbital fibroblasts from GO patients. The addition of caffeine to the adipogenic medium attenuated adipocyte differentiation, as evidenced by microscopic examinations (40×) after Oil Red O staining. The optical density at 490 nm of stained cell lysates also yielded quantitative evidence of the inhibitory effect of caffeine on the accumulation of lipid droplets. (B) Western blot analysis of PPARγ, C/EBPα, and C/EBPβ protein was performed after ten days of adipogenic differentiation. The experiments were performed in duplicate with cells from three different donors. PPARγ, C/EBPα, and C/EBPβ levels determined by densitometry were normalised to the level of β-actin in the same sample. The results are presented as the mean relative density ratios (%) ± standard deviation (*p < 0.05 and **p < 0.01 versus cells without caffeine treatment).

To investigate the signalling pathways through which caffeine mediates its effects in GO orbital fibroblasts, the expression levels of multiple transcription factors were assessed after caffeine treatment. Both GO and non-GO cells were treated with 1 mM caffeine for various durations. Western blot analysis (Fig. 5) showed that caffeine inhibited the phosphorylation of ERK and p38 protein in a time-dependent manner. The phosphorylation of AKT, JNK, and NF-κB remained unchanged with 1 mM caffeine treatment (data not shown).

Fig. 5

Effect of caffeine on intracellular signalling pathways

Orbital fibroblasts from GO patients (n = 3) and non-GO individuals (n = 3) were treated with 1 mM caffeine for 0–180 min. The cell lysates were subjected to western blot analysis. Experiments were performed three times using different strains and samples were assayed in duplicate. Results are expressed as the relative density of each protein normalised to the level of β-actin and presented as means ± standard deviation (*p < 0.05 and **p < 0.01 versus cells without caffeine treatment).

In this study, we found that caffeine curtailed oxidative stress and adipogenesis, the main pathogenic mechanisms of GO. Based on our results, caffeine appears to have an inhibitory effect on ROS generation in response to oxidative stress, as well as an anti-adipogenic effect.

A state of oxidative stress has been described in Graves’ disease and GO [18-20]. In general, an increase in ROS or the reduced elimination of radicals by anti-oxidative enzymes will result in oxidative damage, inflammation, and loss of function. In the in vitro pathogenic mechanism of GO, superoxide radicals stimulate orbital fibroblasts to proliferate and differentiate into mature adipocytes [20]. Based on clinical reports, cigarette smoking—likely the most important environmental factor associated with GO occurrence and progression—may act, among other mechanisms, by stimulating the generation of ROS and reducing antioxidant production [3, 20, 21]. In this regard, several in vitro studies demonstrate the therapeutic effectiveness of drugs with antioxidant potential in primary cultured orbital fibroblasts from GO patients [12, 15-17]. Clinical studies on the use of antioxidants for the management of GO have also shown them to be clinically effective in patients experiencing mild severity [22-24].

Although caffeine is known to act as a free radical scavenger—the basis for its antioxidant activities—it should be noted that it exhibits both anti-oxidant and pro-oxidant properties depending on dose: at low concentrations, caffeine shows antioxidant effects, while at high concentrations, it may increase cytotoxicity, leading to the subsequent generation of reactive oxygen species (ROS) [5, 25, 26]. In this study, however, orbital fibroblasts were resistant to the generation of ROS at high concentrations of caffeine, which acted as an antioxidant in H₂O₂- or CSE-stimulated orbital fibroblasts even at the highest concentration (5 mM) without causing cell damage.

Adipogenesis, the process by which orbital fibroblasts develop into mature adipocytes, is one of the key pathogenic mechanisms in GO. We found that drugs with antioxidant effects significantly suppressed adipocyte differentiation in primary Graves’ orbital fibroblasts [12, 15-17]. Recently, groups studying obesity and metabolism have reported that caffeine inhibits adipogenic differentiation and reduces lipid accumulation [5, 11]. They showed that caffeine may prohibit the activation of C/EBPβ, the first response to adipogenic inducers, blocking the adipogenic cascade involving PPARγ, aP2, or leptin protein. In this study, we showed that ≥0.5 mM caffeine inhibited adipogenic differentiation and lipid accumulation, as well as the expression of adipogenic transcription factors (PPARγ, C/EBPα, and C/EBPβ).

The differentiation of preadipocytes is triggered by hormonal agents and proceeds by the coordination and precise control of transcriptional cascades. The first stage of adipocyte differentiation is growth arrest at a confluent state; proliferating cells will not accumulate lipid droplets in their cytoplasm until they become confluent. Upon induction of differentiation, growth-arrested cells undergo additional rounds of cell division, known as mitotic clonal expansion [27]. A recent study showed that ROS are important in regulating mitotic clonal expansion during adipogenesis and that they facilitate adipocyte differentiation by accelerating mitotic clonal expansion [28]. Moreover, it has been demonstrated that caffeine inhibits adipogenesis through the modulation of mitotic clonal expansion in 3T3-L1 adipocytes [11].

Mitogen-activated protein kinases (MAPKs) constitute a family of protein kinases (ERK, JNK, and the p38-MAPKs in mammals) that play an essential role in relaying extracellular signals from the cell membrane to the nucleus via a cascade of phosphorylation events [29]. MAPKs are key participants in signal transduction pathways activated by mitogenic stimuli, environmental stress, and inflammatory agents [30]. Various studies have documented the involvement of the MAPK signalling pathways in redox-stressed cells and tissues [31]. Moreover, it has been shown that the pathogenesis of GO involves activation of the MAPK signalling pathways, especially the p38 and ERK pathways [32, 33]. Our results showed that the levels of phosphorylated p38 and ERK decreased when they were cotreated at increasing timeframes of caffeine exposure.

The results of the present study show that caffeine has anti-oxidative and anti-adipogenic properties; however, it should be noted that the concentration of caffeine used in this in vitro study is somewhat higher than the normal physiological range. A caffeine concentration of 0.1 mM (19.4 μg/mL) is approximately equivalent to that in a cup of coffee containing 150 mg caffeine (Starbucks^© latte) [5]; it has been reported that caffeine causes adverse effects at excessively high blood concentrations (>80 μg/mL) in humans [34]. In this study, caffeine exhibited anti-adipogenic effects beyond concentrations of 0.5 mM and anti-oxidative effects at 0.1 mM. As GO is a heterogeneous autoimmune disease affected by multiple genetic and environmental factors, in vitro results may not correlate precisely with in vivo conditions. Additionally, we induced pathologic conditions in cells by stimulation using concentrations of chemicals, hormones, and growth factors beyond physiological ranges—for example, our use of oxidants such as H₂O₂ or CSE.

In conclusion, to the best of our knowledge, this is the first study to investigate the effects of caffeine on oxidative stress and adipogenesis in orbital fibroblasts from GO patients. Until now, no clinical or epidemiologic studies have emerged that assess the relationship between caffeine and GD or GO. Our in vitro results provide evidence to support further investigation regarding the influence of caffeine on GO progression in clinical settings.

Funding: This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT (NRF-2017R1A2B4009565).

The authors have no conflicts of interest to declare.

• 1   Lehmann  GM,  Feldon  SE,  Smith  TJ,  Phipps  RP (2008) Immune mechanisms in thyroid eye disease. Thyroid 18: 959–965.
• 2   Burch  HB,  Lahiri  S,  Bahn  RS,  Barnes  S (1997) Superoxide radical production stimulates retroocular fibroblast proliferation in Graves’ ophthalmopathy. Exp Eye Res 65: 311–316.
• 3   Pfeilschifter  J,  Ziegler  R (1996) Smoking and endocrine ophthalmopathy: impact of smoking severity and current vs. lifetime cigarette consumption. Clin Endocrinol (Oxf) 45: 477–481.
• 4   Stan  MN,  Bahn  RS (2010) Risk factors for development or deterioration of Graves’ ophthalmopathy. Thyroid 20: 777–783.
• 5   Su  SH,  Shyu  HW,  Yeh  YT,  Chen  KM,  Yeh  H, et al. (2013) Caffeine inhibits adipogenic differentiation of primary adipose-derived stem cells and bone marrow stromal cells. Toxicol In Vitro 27: 1830–1837.
• 6  Arnaud MJ (1993) Metabolism of caffeine and other components of coffee. In: Garattini S. (ed.), Caffeine, Coffee, and Health. Raven Press, New York, USA: 43–96.
• 7   Dempsey  DA,  Partridge  JC,  Jones  RT,  Rowbotham  MC (1998) Cocaine, nicotine, caffeine, and metabolite plasma concentrations in neonates. J Anal Toxicol 22: 220–224.
• 8   Lv  X,  Chen  Z,  Li  J,  Zhang  L,  Liu  H, et al. (2010) Caffeine protects against alcoholic liver injury by attenuating inflammatory response and oxidative stress. Inflamm Res 59: 635–645.
• 9   Jagdeo  J,  Brody  N (2011) Complementary antioxidant function of caffeine and green tea polyphenols in normal human skin fibroblasts. J Drugs Dermatol 10: 753–761.
• 10   Varma  SD,  Hegde  KR,  Kovtun  S (2010) Oxidative stress in lens in vivo: inhibitory effect of caffeine. A preliminary report. Mol Vis 16: 501–505.
• 11   Kim  AR,  Yoon  BK,  Park  H,  Seok  JW,  Choi  H, et al. (2016) Caffeine inhibits adipogenesis through modulation of mitotic clonal expansion and the AKT/GSK3 pathway in 3T3-L1 adipocytes. BMB Rep 49: 111–115.
• 12   Yoon  JS,  Lee  HJ,  Chae  MK,  Lee  SY,  Lee  EJ (2013) Cigarette smoke extract-induced adipogenesis in Graves’ orbital fibroblasts is inhibited by quercetin via reduction in oxidative stress. J Endocrinol 216: 145–156.
• 13   Kim  SE,  Lee  JH,  Chae  MK,  Lee  EJ,  Yoon  JS (2016) The role of sphingosine-1-phosphate in adipogenesis of Graves’ orbitopathy. Invest Ophthalmol Vis Sci 57: 301–311.
• 14   Ko  J,  Kim  JY,  Lee  EJ,  Yoon  JS (2018) Inhibitory effect of idelalisib, a selective phosphatidylinositol 3-kinase delta inhibitor, on adipogenesis in an in vitro model of Graves’ orbitopathy. Invest Ophthalmol Vis Sci 59: 4477–4485.
• 15   Kim  CY,  Lee  HJ,  Chae  MK,  Byun  JW,  Lee  EJ, et al. (2015) Therapeutic effect of resveratrol on oxidative stress in Graves’ orbitopathy orbital fibroblasts. Invest Ophthalmol Vis Sci 56: 6352–6361.
• 16   Rhiu  S,  Chae  MK,  Lee  EJ,  Lee  JB,  Yoon  JS (2014) Effect of tanshinone IIA in an in vitro model of Graves’ orbitopathy. Invest Ophthalmol Vis Sci 55: 5900–5910.
• 17   Byun  JW,  Hwang  S,  Kang  CW,  Kim  JH,  Chae  MK, et al. (2016) Therapeutic effect of protocatechuic aldehyde in an in vitro model of Graves’ orbitopathy. Invest Ophthalmol Vis Sci 57: 4055–4062.
• 18   Hondur  A,  Konuk  O,  Dincel  AS,  Bilgihan  A,  Unal  M, et al. (2008) Oxidative stress and antioxidant activity in orbital fibroadipose tissue in Graves’ ophthalmopathy. Curr Eye Res 33: 421–427.
• 19   Komosinska-Vassev  K,  Olczyk  K,  Kucharz  EJ,  Marcisz  C,  Winsz-Szczotka  K, et al. (2000) Free radical activity and antioxidant defense mechanisms in patients with hyperthyroidism due to Graves’ disease during therapy. Clinica Chimica Acta 300: 107–117.
• 20   Bartalena  L,  Tanda  ML,  Piantanida  E,  Lai  A (2003) Oxidative stress and Graves’ ophthalmopathy: in vitro studies and therapeutic implications. Biofactors 19: 155–163.
• 21   Shine  B,  Fells  P,  Edwards  OM,  Weetman  AP (1990) Association between Graves’ ophthalmopathy and smoking. Lancet 335: 1261–1263.
• 22   Marcocci  C,  Kahaly  GJ,  Krassas  GE,  Bartalena  L,  Prummel  M, et al. (2011) Selenium and the course of mild Graves’ orbitopathy. N Engl J Med 364: 1920–1931.
• 23   Bednarek  J,  Wysocki  H,  Sowinski  J (2005) Oxidative stress peripheral parameters in Graves’ disease: the effect of methimazole treatment in patients with and without infiltrative ophthalmopathy. Clin Biochem 38: 13–18.
• 24   Bouzas  EA,  Karadimas  P,  Mastorakos  G,  Koutras  DA (2000) Antioxidant agents in the treatment of Graves’ ophthalmopathy. Am J Ophthalmol 129: 618–622.
• 25   Azam  S,  Hadi  N,  Khan  NU,  Hadi  SM (2003) Antioxidant and prooxidant properties of caffeine, theobromine and xanthine. Med Sci Monit 9: BR325–BR330.
• 26   Tiwari  KK,  Chu  C,  Couroucli  X,  Moorthy  B,  Lingappan  K (2014) Differential concentration-specific effects of caffeine on cell viability, oxidative stress, and cell cycle in pulmonary oxygen toxicity in vitro. Biochem Biophys Res Commun 450: 1345–1350.
• 27   MacDougald  OA,  Lane  MD (1995) Transcriptional regulation of gene expression during adipocyte differentiation. Annu Rev Biochem 64: 345–373.
• 28   Lee  H,  Lee  YJ,  Choi  H,  Ko  EH,  Kim  JW (2009) Reactive oxygen species facilitate adipocyte differentiation by accelerating mitotic clonal expansion. J Biol Chem 284: 10601–10609.
• 29   Son  Y,  Cheong  YK,  Kim  NH,  Chung  HT,  Kang  DG, et al. (2011) Mitogen-activated protein kinases and reactive oxygen species: how can ROS activate MAPK pathways? J Signal Transduct 2011: 792639.
• 30   Kyriakis  JM,  Avruch  J (2012) Mammalian MAPK signal transduction pathways activated by stress and inflammation: a 10-year update. Physiol Rev 92: 689–737.
• 31   Gaitanaki  C,  Konstantina  S,  Chrysa  S,  Beis  I (2003) Oxidative stress stimulates multiple MAPK signalling pathways and phosphorylation of the small HSP27 in the perfused amphibian heart. J Exp Biol 206: 2759–2769.
• 32   Chen  B,  Tsui  S,  Smith  TJ (2005) IL-1 beta induces IL-6 expression in human orbital fibroblasts: identification of an anatomic-site specific phenotypic attribute relevant to thyroid-associated ophthalmopathy. J Immunol 175: 1310–1319.
• 33   Kazim  M,  Goldberg  RA,  Smith  TJ (2002) Insights into the pathogenesis of thyroid-associated orbitopathy: evolving rationale for therapy. Arch Ophthalmol 120: 380–386.
• 34   Nomura  M,  Ichimatsu  D,  Moritani  S,  Koyama  I,  Dong  Z, et al. (2005) Inhibition of epidermal growth factor-induced cell transformation and Akt activation by caffeine. Mol Carcinog 44: 67–76.