Possible implications of TGF-alpha in oesophageal dysmotility in systemic sclerosis M. Lalovac1, D. Martinovic-Kaliterna2, S. Mejic Krstulovic1, V. Markovic3, I. Salamunic4, D. Perkovic2

1Department of Internal Medicine, ABSTRACT broblasts, lymphocyte activation, au- County Hospital Dubrovnik, Croatia; Objective. Hypoxia is a characteris- toantibody production, inflammation, 2 Department of Rheumatology and tic feature of systemic sclerosis (SSc). and tissue fibrosis must be considered Clinical Immunology, 3Department of Transforming growth factor alpha (1). Fibroblast activation leads to fibro- Nuclear Medicine, 4Department of Laboratory Diagnostics, University (TGF-α) has an important role in ex- genesis and accumulation of extracel- Hospital Center Split, Croatia. cessive inflammation under hypoxic lular matrix (ECM) which modulates Milos Lalovac, MD conditions. Since oesophageal dysmo- the activity of mediators such as TGFß Dusanka Martinovic-Kaliterna, PhD tility is one of the most common signs and TGF-α (2). TGF-α is a member of Stanka Mejic Krstulovic, MD of SSc, the aim of this study was to ex- epidermal growth factor (EGF) family Vinko Markovic, PhD plore the relation between TGF-α and of cytokines. It is a small mitogenic Ilza Salamunic, PhD oesophageal dysmotility in SSc. protein that consists of 50 aminoac- Dijana Perkovic, MD Methods. The study included 35 pa- ids linked with 3 disulfide bonds (3). Please address correspondence to: tients with SSc and 32 healthy controls It is excreted by macrophages, brain Milos Lalovac, matched for sex and age. Serum con- cells, keratinocytes and fibroblasts (4). Department of Internal Medicine, centrations of TGF-α were measured County Hospital Dubrovnik, In rat cell cultures, TGF-α stimulates R. Misetica bb, using ELISA. Oesophageal motility was the growth of normal fibroblasts (5). HR-20000 Dubrovnik, Croatia. assessed by oesophageal scintigraphy. For this effect, TGFα requires TGFβ, E-mail: [email protected] A multiple-swallow test was performed which augments activation of TGF-α 99m Received on February 5, 2014; accepted in the study population with Tc-DT- through a separate receptor system (6). in revised form on April 28, 2014. PA. A region of interest over the entire Bound to epidermal growth factor re- Clin Exp Rheumatol 2014; 32 (Suppl. 86): oesophagus was defined and the reten- ceptor (EGFR) TGF-α causes recep- S149-S153. tion index (RI) was calculated. tor dimerisation, activation of tyrosine © Copyright Clinical and Results. Statistically significant differ- kinase, and activation of a signaling Experimental Rheumatology 2014. ences in serum concentration of TGF-α cascade (7). EGFR are expressed by as well as of RI of 99mTc-DTPA were many types of cells including skin Key words: systemic sclerosis, found between patients with SSc and keratinocytes, fibroblasts, vascular hypoxia, TGF-α, oesophageal healthy controls. A statistically signifi- endothelial cells, and epithelial cells scintigraphy, 99mTc-DTPA cant correlation was found between se- of the gastrointestinal tract (GIT) (8). rum concentrations of TGF-α and RIs At the cellular level, EGFR stimulates of 99mTc-DTPA. This correlation was cell proliferation and also can stimu- inverse, i.e. when serum concentrations late angiogenesis (9-10). TGF-α has of TGF-α increased, the RI of 99mTc-DT- an important role in the stimulation of PA decreased (Spearman rho =-0361, excessive inflammation in a hypoxic p=0.033). environment (11). Hypoxia is an initial Conclusion. These results point to a event of ischaemic and lypopolisacha- possible relation between TGF-α and ride stimulated inflammation (12). It is oesophageal dysmotility in SSc. Al- also a strong inductor of TGF-α tran- though the results do not explain the ex- scription and translation as well as a act role of this cytokine in the pathogen- crucial mediator of the generalised sys- esis of esophageal changes, the finding temic inflammatory response (13-14). of inverse correlation between TGF-α TGF-α has been extensively studied on and oesophageal dysmotility is intrigu- acute and chronic injury models of the ing and requires further investigation. upper gastrointestinal mucosa. During regeneration of upper gastrointestinal Introduction mucosal injury TGF-α and EGF serve When discussing the pathogenesis predominantly to restore the epithelial of SSc, communication between en- component (15). TGF-α has been stud- Competing interests: none declared. dothelial cells, epithelial cells and fi- ied also on an interstitial lung disease

S-149 TGF-α and oesophageal dysmotility in SSc / M. Lalovac et al. model with findings that lung epithe- nant disease, 7 patients had problems Measurement of TGF-α lial-specific overexpression of TGF-α with upper gastrointestinal tract other All subjects were tested for TGF-α. Se- leads to progressive and pronounced than related to SSc. Four patients did rum samples were taken via vein punc- pulmonary fibrotic lesions in transgen- not wish to participate in the study. Due ture in the morning, after an overnight ic mice (16-18). On a kidney fibrosis to a low incidence of disease and lack fast, and collected in 2 blood tubes. model, interaction of EGF and TGFβ of scientific work on the role of TGF-α Serum was extracted and stored in a has been documented (19-20). The role in SSc, we decided to enroll only pa- freezer (temperature: -70oC). TGF-α of TGFβ in SSc has been studied in de- tients with dcSSc (largest homogenous concentrations were measured in dupli- tail, but the effects of members of the group of patients) and to perform the cate, using ELISA method on an auto- epidermal growth factor family, such study on such a homogenous sample. matic analyser miniBOS, manufactured as TGF-α, have not been studied yet The control group consisted of healthy by Biomedica (Automatic microplate (21-22). Clinically, aside from the skin, volunteers and outpatients suffering analyser, Biomedica Gruppe, Wien, in SSc patients, the digestive tract is from cervico-brachial and lumbosacral Austria). Polyclonal antibody specific also frequently affected – with the oe- syndrome having visited our immunol- for TGFα were pre-coated onto mi- sophagus being involved in 75–90% of ogy outpatient clinic during the period croplates. Standards and samples were patients with SSc (23). Excessive fibro- of recruitment. Controls were free of pipetted into the wells and any TGFα sis, inflammation, and vascular - dam any problems with upper GIT, having present was bound by the immobilised age are basic processes of GIT damage normal endoscopic findings of the distal antibody. After washing away any un- with a clinical presentation of dysfunc- part of oesophagus within 6 months of bound substances, an enzyme-linked tion pertaining to motility, digestion, the time of recruitment. Disease dura- polyclonal antibody specific for TGF-α absorption, or excretion (24). Typical tion ranged from 18 to 60 months, with was added to the wells. Following a SSc changes include atrophy of smooth a mean of 46 months. SSc was diag- wash to remove any unbound antibody- oesophageal muscle in the distal 2/3 of nosed according to the revised criteria enzyme reagent, a substrate solution oesophagus with consequent reflux oe- of the American Rheumatology Associ- was added to the wells and colour de- sophagitis and stenosis, strictures (17- ation (29). All of 35 patients had diffuse veloped in proportion to the amount of 29%), and Barrett’s oesophagus (0- cutaneous SSc, underwent a complete TGF-α bound in the initial step. The 37%) (25-26). Oesophageal involve- physical examination, and were ques- colour development was stopped and ment is not necessarily symptomatic, tioned about possible malignant diseas- the intensity of the colour was meas- therefore it may be diagnosed late in es in their past and any family history ured. A standard curve was included on the course of the disease, when compli- of malignant diseases in first-degree each plate using doubling dilutions of cations have already set in (27). There- relatives before the age of 65. Age, sex pure recombinant TGF-α (R and D sys- fore, we decided to investigate the rela- and current treatment were recorded. tems). Data was processed using GraFit tion between TGF-α and oesophageal Routine haematological and biochemi- software (Microsoft). Results were ex- changes. Oesophageal motility was cal measurements were performed. All pressed in pg/ml (Quantikine, Human assessed by oesophageal scintigraphy subjects were tested for ACA-Cenp TGF-α immunoassay, DTGA00, R&D because it is less invasive and is easily B, anti-Scl-70 and anti-RNP antibod- Systems Inc., Minneapolis, USA) (30). performed for patients than manometry ies from analysis of multiple measure- or oesophagoscopy (28). ments of autoantibodies (AtheNA Multi Measurement of RI of 99mTc-DTPA – Lyte ANA test system Zeus Scientific, Oesophageal motility was determined Materials and methods Inc.Raritan, NY, USA) using an adjust- by oesophageal scintigraphy with tech- Study population ed flow cytometer Luminex 100 (Aus- netium 99 m dietilentriaminopentaceta- The study included 35 consecutive pa- tin, USA). Results were expressed in ta (Tc-99m-DTPA). The percentage of tients with SSc classified according to U/ml. Negative (normal) values ranged Tc-99m-DTPA was used as the measure EUSTAR (European Scleroderma Trial up to 40 U/ml, whereas positive (high) of (RI). The level of 100% was consid- and Research) criteria and 32 control values were greater than 40 U/ml. Ex- ered as highest oesophageal dysfunc- subjects matched for age (± 3 years) and clusion criteria were as follows: previ- tion, and levels below 40% were con- sex. Patient recruitment was performed ous or current malignant disease. All sidered normal. The test was performed by three licensed rheumatologists patients (with SSc or controls) with hia- in the afternoon hours, after subjects through the Department of Rheuma- tus hernia, oesophageal erosions, ulcers fasted for 5 hours. In a supine position, tology, University Hospital Split, Split or gastroesophageal reflux disease, H. through a 20 ml syringe, a subject was Croatia in the time period from Janu- pylori infection were excluded. Written given 300 μCi Tc-99m-DTPA in 3 ml of ary 2011 to January 2012. Of the 65 ex- informed consent (according to the Hel- water per orally (i.e. to swallow). Every amined SSc patients, 35 patients with sinki Declaration) was obtained from 15 seconds subject was asked to swal- diffuse cutaneous SSc (dcSSc) were all study subjects prior to study par- low again, and for 2 minutes dynamics enrolled in our study. Thirteen of these ticipation. The Ethics Committee of the of passage of radioactive bolus through patients had a limited cutaneous form Split University Hospital Center, Split, the oesophagus, and oesophageal emp- of disease, 6 patients had active malig- Croatia approved the study protocol. tying activities were recorded. RI was

S-150 TGF-α and oesophageal dysmotility in SSc / M. Lalovac et al. measured at 15, 30, 60 and 90 seconds. Table I. Difference in TGF-α level between SSc patients and control group. RI was shown as a percentage and the most significant results were measured SSc group Control group at 60 seconds. These values were used median (IQR) median (IQR) p ; effect for analysis purposes. Measurements were done using a Siemens Orbiter TGF-α 16 (8–33.6) 10.3 (7.1–13.6) 0.044; 0.36 Gamma Camera and a PDP 11/34 com- IQR: interquartile range; *p<0.05 puter, 11 Gamma software package. Table II. Difference in RI of 99mTc-DTPA between SSc patients and control group. Statistical analysis The level of significance was set at SSc group Control group 5% (p<0.05), and all confidence -in tervals were given on the 95% level. median (IQR) median (IQR) p ; effect Normality of distribution of continu- RI of 99mTc-DTPA 94 (52–100) 25 (17–37.3) 0.001*; 0.14 ous variables like TGFα was tested by Kolmogorov-Smirnov test for samples IQR: interquartile range; *p<0.05 greater than 30, or by Shapiro-Wilk test 99m for samples smaller than 30. Median Table III. Relation of TGF-α and oesophageal dysmotility (RI of Tc-DTPA). and interquartile ranges were used as n Spearman rho p-value measures of central tendency and vari- ability when the distribution statisti- Study population 67 0.037 0.740 * cally significantly deviated from the SSc group 35 -0.361 0.033 Control group 32 0.344 0.054 normal one. Differences in mean values for the continuous, numeric variables *p<0.05 between two categories of a nomi- nal variable were analysed by Mann- tral tendency and variability for both This correlation was inverse, meaning Whitney U-test. AUC was given as a groups. No statistically significant dif- that with an increase in TGF-α, RI of standardised measure of the effect size ferences regarding gender (χ2=0.013; 99mTc-DTPA decreased (Table III). for statistically significant results, and p>0.999), nor age (Mann-Whitney There was no statistically significant re- is calculated according to the formula: U=179.5; Z=-0.176; p=0.871) were lation between serum concentration of U / (m * n), where U is the result of found between the control subjects and TGF-α and any of specific antibodies the Mann-Whitney test (rank sum), m the SSc patients. We also measured se- (anti-Scl-70 - Spearman’s rho -0.301; and n are sizes of two samples. Pos- rum concentration of autoantibodies p=0.079; ACA Cent B - Spearman rho sible impact of covariates on relation- (anti-Scl-70, ACA Cenp B and anti- -0.218; p=0.207; anti-RNP - Spear- ship between TGF-α level and systemic RNP). In SSc patients group 12 patients man‘s rho 0.254; p=0.141). sclerosis was controlled by ANCOVA. had anti-Scl-70 antibodies (34%), 7 There was no statistically significant Differences between two nominal vari- patients had ACA Cenp B antibodies correlation between serum concentra- ables were analysed by chi-square test (20%) and 6 patients had anti-RNP an- tion of TGF-α and Rodnan skin score with, and Cramer’s V was given as a tibodies (17%) in serum concentrations in SSc patients group (Spearman rho = standardised measure of the effect size of more than 40 U/ml. Sera of 10 SSc 0.033; p=0.851). for statistically significant results. Re- patients were negative for these three Also, no statistically significant differ- lationship between two continuous, nu- antibodies (29%). Also, the sera of the ence was found regarding serum con- meric variables was analysed by Spear- control subjects were negative for these centration of TGF-α in SSc patients man’s rho coefficient. All analyses were three antibodies (100%). with pulmonary involvement and SSc carried out using SPSS 17.0 (SPSS Inc., SSc patients had statistically significant- patients without signs of pulmonary in- Chicago, IL, USA) statistical software ly higher average values of TGFα than volvement (Z=-0.454, p=0.650). package. subjects in the control group (Mann - Whitney U=400.5; Z=-2.003; p=0.044; Discussion Results AUC=0.36) (Table I). SSc patients had In our study, we demonstrated that The study population consisted of 35 statistically significantly higher average serum concentrations of TGF-α in pa- (52.2%) SSc patients and 32 (47.8%) values of RI of 99mTc-DTPA than con- tients with SSc were statistically sig- control subjects. As the distribution of trol group (Mann-Whitney U=156; Z=- nificantly higher in comparison to con- age significantly statistically deviated 5.101; p<0.0001; AUC=0.14) (Table trol group. This finding could implicate from normal in the group of SSc pa- II). Statistically significant correlation possible relation between TGF-α and tients (Kolmogorov-Smirnov z=0.16; was found between level of TGF-α and inflammatory processes in SSc, that p=0.040), the median and interquartile value of RI of 99mTc-DTPA in SSc pa- have an impact on fibroblast activation ranges were used as measures of cen- tients (Spearman rho=-0.361; p=0.033). and fibrogenesis in SSc.

S-151 TGF-α and oesophageal dysmotility in SSc / M. Lalovac et al.

Why did we decide to investigate cesses, in our study, serum TGF-α con- Normally, not only macrophages and TGF-α concentrations in the sera of cetrations were analysed in subgroup of keratinocytes, but also epithelial cells SSc patients? To date, there are neither SSc patients, comparing those with pul- of GIT synthesise and secrete TGF-α. sensitive nor specific laboratory tests monary involvement and those without Presumably, this secretion would be for the diagnosing of patients with SSc. pulmonary involvement. No statistical higher in an inflammatory environment The main diagnostic criteria for SSc are significant difference was found. of esophagus at the beginning of oe- performed almost exclusively by clini- Furthermore, TGF-α and EGF have al- sophageal involvement. cal assessment of typical localised skin ready been identified in GIT of humans A potential limitation of our study lesions, but also of fibrotic changes to and of rats where they act as potent was that the sample size was not large internal organs (31). Antinuclear anti- mitogens for certain types of cells. Be- enough due to the low incidence of bodies are found in more than 90% of sides this mitogenic effect, members of disease. This small sample was further SSc patients, while the more specific the EGF cytokine family can modulate subdivided into subgroups which relat- anti-Scl-70 antibody (that could serve cell migration, mucus production and ed to organ involvement, namely pul- as serological marker of SSc) is found secretion, gastric acid secretion and monary involvement. Analysis of such in only 30–40% of these patients (32). gastrointestinal motility (38). Scheiman a small subgroup of patients is not sta- A major pathogenic hallmark of disease et al. documented the relation between tistically meaningful. Also, this study is the fibroproliferative process, not local TGF-α concentrations in gastric was cross-sectional but, considering the only restricted to the skin, but also very body, antrum and duodenum and medi- chronic nature of this disease and vari- prominent in the lungs, heart and gas- camentous therapy, namely with PPI ous rate of progression, it would be in- trointestinal tract (33). Expression of and acetylsalicylic acid (39). This study teresting to perform a large prospective TGFβ receptors is increased in SSc fi- was performed on biopsy specimens study of serum TGF-α concentrations broblasts, and increased concentrations measuring local gastric and duodenal at regular time intervals but also relat- of TGFβ is found in the lungs of pa- levels of TGF-α, while our study meas- ed to changes in oesophageal function tients with idiopathic pulmonary fibro- ured serum concentrations of TGF-α. through the course of the disease in the sis (34). There is strong evidence that Out of 65 patients treated in our facility same patient. It is a promising concept, EGF ligands and receptors are critical during the period of recruitment, 7 pa- but further investigation is required. mediators in the fibrogenic responses to tients were taking high dosages of PPI TGFβ. EGFR and ligands are found to due to different pathological conditions References be elevated in various interstitial lung in the upper gastrointestinal tract, but 1. VARGA J: Systemic sclerosis: an update. Bull NYU Hosp Jt Dis 2008; 66: 198-202. diseases (35). Also, inhibition of TGFβ due to these pathological endoscopic 2. VARGA J, ABRAHAM D: Systemic sclerosis: on the animal pulmonary fibrosis model findings they were excluded from the a prototypic multisystem fibrotic disorder. mitigates fibroproliferative changes in study during the recruitment process. J Clin Invest 2007; 117: 557-67. the lungs. In addition to the lung, mod- One enrolled patient was taking acetyl- 3. MOY FJ, LI YC, RAUENBUEHLER P, WIN- KLER ME, SCHERAGA HA, MONTELIONE els of renal fibrosis also demonstrate salicylic acid due to comorbid condi- GT: Solution structure of human type-alpha fibrosis progression despite inhibition tions (coronary artery disease). Endo- transforming growth factor determined by of the TGFβ pathway (36, 19). These scopic finding of this subject’s upper heteronuclear NMR spectroscopy and re- fined by energy minimization with restraints. findings point out that both EGFR and gastrointestinal tract showed no GERD, Biochemistry 1993; 32: 7334-53. TGFβ pathways have a central role in gastric erosions nor ulcers, and he was 4. BENNETT NT, SCHULTZ GS: Growth factors the pathogenesis of fibroproliferative therefore enrolled in our study. and wound healing: Biochemical properties processes (37). To date, TGF-α, the In our study, oesophageal scintigraphy of growth factors and their receptors. Am J 99m Surg 1993; 165: 728-37. main EGFR ligand, was not studied in was performed and RI of Tc-DTPA 5. KOYAMA S, PODOLSKY DK: Differential SSc patients. was used as a measure of oesophageal expression of transforming growth factors α During the recruitment period, 35 dc- dysfunction. Oesophageal scintigraphy and β in rat intestinal epithelial cells. J Clin SSc patients were enrolled in the study was chosen because it is a safe method Invest 1989; 83: 1768-73. 6. RUSCH V, MENDELSOHN J, DMITROVSKI E: group, therefore presenting the largest that causes little discomfort to patients The epidermal growth factor receptor and its homogenous group of patients. Accord- and is better tolerated than manometry. ligands as therapeutic targets in human tu- ing to pulmonary involvement, a sub- At the same time, it shows great ac- mors. Cytokine Growth Factor Rew 1996; 7: 133-41. group of 7 patients with signs of pul- curacy in diagnosing non-obstructive 7. BEYER C, DISTLER JHW, DISTLER O: Are monary involvement was formed. All oesophageal dysphagia (40). It was su- tyrosine kinase inhibitors promising for the 7 patients had signs of lung fibrosis on prising to find an inverse correlation be- treatment of systemic sclerosis and other fi- HRCT, five patients had significantly re- tween serum concentrations of TGF-α brotic diseases? Swiss Med Wkly 2010; 14: 99m 13050. duced DLCO, and two patients had both and value of RI of Tc-DTPA in SSc 8. SCHULTZ G, CLARK W, ROTATORI DS: lung fibrosis and pulmonary arterial hy- patients. Perhaps, this could be due to EGF and TGF-α in wound and repair. J Cell pertension. Since majority of previous loss of cellular TGF-α secreting ele- Biochem 1991; 45: 346-52. works regarding TGF-α in fibroprolifer- ments during pathologic collagen dep- 9. De LUCA A, CAROTENUTO A, RACHIGLIO A et al.: The role of the EGFR signaling in tu- ative processes concentrated on the role osition and fibrous overgrowth in oe- mor microenvironment. J Cell Physiol 2008; of TGF-α on pulmonary fibrotic pro- sophagus in the natural course of SSc. 214: 559-67.

S-152 TGF-α and oesophageal dysmotility in SSc / M. Lalovac et al.

10. CRUIJSEN VH, GIACCONE G, HOEKMAN K: 20. LAN HY: Diverse roles of TGF-B/SMADS Mod Rheumatol 2013; 23: 1076-84. Epidermal growth factor receptor and angio- in renal fibrosis and inflammation. Int J Biol 32. SAVAS N, DAGLI U, ERTUGRUL E, KURAN S, genesis: opportunities for combined antican- SCi 2011; 7: 1056-67. SAHIN B: Autoantibody profile in Systemic cer strategies. Int J Cancer 2006; 118: 883-8. 21. BLOBE GC, SCHIEMANN WP, LODISCH HF: sclerosis as a marker for esophageal and oth- 11. RAPPOLEE DA, MARK D, BANDA MJ, WERB Role of transforming growth factor beta in er organ involvement in Turkish populations. Z: Wound macrophages express TGF-alpha human disease. N Engl J Med 2000; 342: Dig Dis Sci 2007; 52: 3081-6. and other growth factors in vivo: analysis 1350-8. 33. KAWAKAMI T, IHN H, XU W, SMITH E, LEROY by mRNA phenotyping. Science 1988; 241: 22. COTTON SA, HERRICK AL, JAYSON MI, FREE- C, TROJANOWSKA M: Increased expression 708-12. MONT AJ: TGF beta – a role in systemic scle- of TGF-beta receptors by scleroderma fibro- 12. JANTSCH J, CHAKRAWORITTY D, TURZA N rosis?. J Pathol 1998; 184: 4-6. blasts: evidence for contribution of autocrine et al.: Hypoxia and hypoxia-inducible factor- 23. MARIE I: Gastrointestinal involvement in TGF-beta signaling to scleroderma pheno- 1α modulate lipopolysaccharide-induced systemic sclerosis. Presse Med 2006; 35: type. J Invest Derm 1998; 110: 47-51. dendritic cell activation and function. J Im- 1952-65 34. KHALIL N, PAREKH TV, OCONNOR R et al.: munol 2008; 180: 4697-705. 24. EBERT EC: Esophageal disease in sclero- Regulation of the effects of TGF-beta 1 by 13. GUNARATNAM L, MORLEY M, FRANOVIC A derma. J Clin Gastroenterol 2006; 40: 769-75. activation of latent TGF-beta 1 and differen- et al.: Hypoxia inducible factor activates the 25. NTOUMAZIOS SK, VOULGARI PV, POTSIS K, tial expression of TGF-beta receptors (T beta transforming growth factor-alpha/epidermal KOUTIS E, TSIFETAKI N, ASSIMAKOPOULOS R-I and T beta R-II) in idiopathic pulmonary growth factor receptor growth stimulatory DA: Esophageal involvement in scleroderma: fibrosis. Thorax 2001; 56: 907-15. pathway in VHL (-/-) renal cell carcinoma gastroesophageal reflux, the common -prob 35. BAUGHMAN RP, LOWER EE, MILLER MA, cells. J Biol Chem 2003; 45: 44966-74. lem. Semin Arthritis Rheum 2006; 36: 173- BEJARANO PA, HEFFELFINGER SC: Overex- 14. WANG D, MALO D, HEKIMI S: Elevated mito- 81. pression of transforming growth factor-alpha chondrial reactive oxygen species generation 26. DOMSIC R, FASANELLA K, BIELEFELDT K: and epidermal growth factor-receptor in idi- affects the immune response via hypoxia- Gastrointestinal manifestations of systemic opathic pulmonary fibrosis.Sarcoidosis Vasc inducible factor-1 α in long-lived Mclk1 ± sclerosis. Dig Dis Sci 2008; 53: 1163-74. Diffuse Lung Dis 1999; 16: 57-61. mouse mutants. J Immunol 2010; 184: 582-90. 27. WOJAS-PELC A, STEPIEN A, LIPKO-GOD- 36. ISHII Y, FUJIMOTO S, FUKUDA T: Gefitinib 15. JONES MK, TOMIKAWA M, MOHAJER B, LEWSKA S, WIELOWIEYSKA-SZYBINSKA D, prevents bleomycin-induced lung fibrosisin TARNAWSKI AS: Gastrointestinal mucosal ZABINSKA-PLAZAK E, KIELTYKA A: Esoph- mice. Am J Respir Crit Care Med 2006; 174: regeneration: role of growth factors. Front ageal scintigraphy in patients with systemic 550-6. Biosci 1999; 4: D303-9. sclerosis: clinical symptoms correlated with 37. MADALA SK, KORFHAGEN TR, SCHMIDT S 16. BURGEL PR, NADEL JA: Epidermal growth the esophagus noted by the patients. Przegl et al.: Inhibition of the αvβ6 integrin leads factor receptor-mediated innate immune re- Lek 2002; 59: 973-6. to limited alteration of TGFα-induced pul- sponses and their roles in airway diseases. 28. MARIANI G, BONI G, BARRECA M et al.: monary fibrosis.Am J Physiol Lung Cell Mol Eur Respir J 2008; 32: 1068-81. Radionuclide gastroesophageal motor stud- Physiol 2014; 306: L726-35. 17. WAHEED S, DANGIO CT, WAGNER CL et ies. J Nucl Med 2004; 45: 1004-28. 38. HOFFMANN P, ZEEH JM, LAKSHMANAN J al.: Transforming growth factor alpha 29. Subcommittee for scleroderma criteria of the et al.: Increased expression of transforming (TGF(alpha)) is increased during hyperoxia American Rheumatism association dignostic growth factor α precursors in acute experi- and fibrosis.Exp Lung Res 2002; 28: 361-72. and therapeutic criteria committee. Prelimi- mental colitis in rats. Gut 1997; 41: 195-202. 18. KORFHAGEN TR, SWANTZ RJ, WERT SE et nary criteria for the classification of systemic 39. SCHEIMAN JM, MEISE KS, GREENSON JK, al.: Respiratory epithelial cell expression of sclerosis. Arthritis Rheum 1980; 23: 581-90. COFFEY RJ: Transforming growth factor-al- human transforming growth factor-alpha in- 30. CATALOG NUMBER DTGAOO: For the quan- pha (TGF-α) levels in human proximal gas- duces lung fibrosis in transgenic mice.J Clin titative determination of human transforming trointestinal epithelium. Effect of mucosal Invest 1994; 93: 1691-9. growth factor alpha (TGF-α) concentrations injury and acid inhibition. Dig Dis Sci 1997; 19. CHUNG H, RAMACHANDRAN R, HOLLEN- in cell culture supernates, serum, plasma, and 42: 333-41. BERG MD, MURUVE DA: Proteinase-activat- human milk. Human TGF-α Immunoassay- 40. PARKMAN HP, MAURER AH, CAROLINE DF, ed receptor-2 transactivation of epidermal Quantikine 2010 RD Systems. MILLER DL, KREVSKY B, FISHER RS: Opti- growth factor receptor and transforming 31. HASEGAWA M, ASANO Y, ENDO H et al.: mal evaluation of patients with nonobstruc- growth factor- β receptor signaling pathways Serum chemokine levels as prognostic mark- tive esophageal dysphagia. Manometry, scin- contributes to renal fibrosis. J Biol Chem ers in patients with early systemic sclerosis: tigraphy, or videoesophagography? Dig Dis 2013; 52: 37319-31. a multicenter, prospective, observation study. Sci 1996; 41: 1355-68.

S-153