uncontrolled expression of vascular endothelial growth ... fileuncontrolled expression of vascular...

Gassmann and Steffen GayBeat A. Michel, Renate E. Gay, Ulf Müller-Ladner, Marco Matucci-Cerinic, Karl H. Plate, Max

Oliver Distler, Jörg H. W. Distler, Annette Scheid, Till Acker, Astrid Hirth, Janine Rethage,to Insufficient Skin Angiogenesis in Patients With Systemic Sclerosis

Uncontrolled Expression of Vascular Endothelial Growth Factor and Its Receptors Leads

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 2004 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.0000134644.89917.962004;95:109-116; originally published online June 3, 2004;Circ Res.

http://circres.ahajournals.org/content/95/1/109World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circres.ahajournals.org/content/suppl/2004/07/04/95.1.109.DC1.htmlData Supplement (unedited) at:

http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Researchin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on February 23, 2013http://circres.ahajournals.org/Downloaded from

http://circres.ahajournals.org/content/95/1/109

http://circres.ahajournals.org/content/suppl/2004/07/04/95.1.109.DC1.html

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circres.ahajournals.org//subscriptions/

http://circres.ahajournals.org/

Uncontrolled Expression of Vascular Endothelial GrowthFactor and Its Receptors Leads to Insufficient Skin

Angiogenesis in Patients With Systemic SclerosisOliver Distler, Jorg H.W. Distler, Annette Scheid, Till Acker, Astrid Hirth, Janine Rethage,Beat A. Michel, Renate E. Gay, Ulf Muller-Ladner, Marco Matucci-Cerinic, Karl H. Plate,

Max Gassmann, Steffen Gay

Abstract—Systemic sclerosis (SSc) skin lesions are characterized by disturbed vessel morphology with enlarged capillariesand an overall reduction in capillary density, suggesting a deregulated, insufficient angiogenic response. It has beenpostulated that this phenomenon is due to reduced expression of the potent angiogenic factor vascular endothelial growthfactor (VEGF). In contrast to this hypothesis, we demonstrate that the expression of both VEGF and its receptorsVEGFR-1 and VEGFR-2 is dramatically upregulated in skin specimens of SSc patients throughout different diseasestages. Interestingly, upregulation of VEGF was not mediated by hypoxia-inducible transcription factor-1 (HIF-1) asindicated by only a weak expression of the oxygen-sensitive �-subunit of HIF-1 in the skin of SSc patients. This wasunexpected on measuring low PO2 values in the SSc skin by using a polarographic oxygen microelectrode system.Considering our observation that PDGF and IL-1� costimulated VEGF expression, we propose that chronic anduncontrolled VEGF upregulation that is mediated by an orchestrated expression of cytokines rather than VEGFdownregulation is the cause of the disturbed vessel morphology in the skin of SSc patients. Consequently, for therapeuticapproaches aiming to improve tissue perfusion in these patients, a controlled expression and timely termination of VEGFsignaling appears to be crucial for success of proangiogenic therapies. (Circ Res. 2004;95:109-116.)

Key Words: angiogenesis � vascular endothelial growth factor � hypoxia-inducible transcription factor-1

Systemic sclerosis (SSc) is a multiorgan disease charac-terized by widespread fibrosis, activation of immune

cells, production of autoantibodies, and injury to vascular aswell as microvascular structures.1 The earliest clinical symp-toms of SSc relate to disturbances in the peripheral vascularsystem.2 Nailfold capillaroscopy shows a variety of morpho-logical changes including enlarged capillaries, bushy capil-lary formations, microhemorrhages, and a variable loss ofcapillaries with or without avascular areas. These phenomenaare often accompanied by increased markers of endothelialcell injury and endothelial cell activation.3

Despite the reduced capillary density, there is paradoxi-cally no sufficient angiogenic response in the skin of patientswith SSc.4 Tissue ischemia leads usually to the expression ofangiogenic growth factors, which then initiate angiogenicsprouting by inducing vasodilatation, proliferation, and mi-gration of endothelial cells and stabilization of the lumina toform new vessels.5

Among the angiogenic growth factors, vascular endothelialgrowth factor (VEGF) has been identified as a key mediator

of angiogenesis. VEGF induces differentiation, proliferation,and migration of endothelial cells that contribute to theformation of vessels through both angiogenesis and vascularremodeling. VEGF exerts its biological functions by bindingto the tyrosine kinase receptors VEGFR1 (Flt-1) andVEGFR2 (Flk-1). Deletion of VEGF as well as of itsreceptors lead to embryonic lethality due to vascular defects,further highlighting the complex and multiple functions of theVEGF/VEGF receptor axis for angiogenic processes (seereview)5. The expression of VEGF is tightly controlled andstimulated under hypoxic conditions through binding of thetranscription factor hypoxia-inducible transcription factor-1(HIF-1), a heterodimeric transcription factor consisting of theoxygen-regulated �-subunit and the constitutively expressed�-subunit, to a 47-bp sequence located 985 to 939 bpupstream of the transcription initiation site.6

Considering the lack of a sufficient angiogenesis despitethe reduction in the capillary density, we hypothesized thatthe expression of the potent proangiogenic growth factor

Original received December 1, 2003; revision received May 21, 2004; accepted May 24, 2004.From the Center of Experimental Rheumatology, Department of Rheumatology (O.D., J.H.W.D., A.H., J.R., B.A.M., R.E.G., S.G.), University Hospital

Zurich, Switzerland; Institute of Physiology (A.S.), University of Zurich, Switzerland; Karolinska Institute (T.A.), Stockholm, Sweden; Institute ofNeurology (Edinger Institute) (T.A., K.H.P.), Johann-Wolfgang Goethe University, Frankfurt, Germany; Department of Internal Medicine I (U.M.-L.),University of Regensburg, Germany; Department of Medicine (M.M.-C.), Section of Rheumatology, University of Florence, Italy; and the Institute ofVeterinary Physiology (M.G.), University of Zurich, Switzerland.

Correspondence to Oliver Distler, MD, Center of Experimental Rheumatology, Department of Rheumatology, University Hospital Zurich, CH-8091Zurich, Switzerland. E-mail [email protected]

© 2004 American Heart Association, Inc.

Circulation Research is available at http://www.circresaha.org DOI: 10.1161/01.RES.0000134644.89917.96

109

Clinical Research



VEGF might be downregulated in patients with SSc. In thepresent study, we therefore analyzed the expression andregulation of VEGF and its receptors in skin specimens fromSSc patients. In addition, we addressed the impact of HIF-1�on VEGF expression in SSc patients.

Materials and MethodsFor an expanded Materials and Methods, see the online datasupplement available at http://circres.ahajournals.org. Characteristicsof SSc patients are shown in online Table 1.

Patients and Skin BiopsiesSkin biopsies were obtained from 15 patients who met the AmericanCollege of Rheumatology criteria for SSc.7 In all patients, biopsieswere taken from clinically involved skin. In a subset of patients(n�7), biopsies were also taken from clinically noninvolved skin asassessed by skin scoring of an experienced examiner. Controls (n�6)consisted of biopsies from healthy volunteers. Experiments wereapproved by local ethics review committees, and written informedconsent was obtained from all patients.

In Situ HybridizationPlasmids containing the 517 bp human VEGF121 cDNA were kindlyprovided by H.A. Weich (Max-Planck-Institute, Bad Nauheim,Germany) and used as a template to design VEGF-A–specificriboprobes.8 Preparation of VEGF-A–specific riboprobes and non-radioactive in situ hybridization was performed as described else-where.9 Hybridized probes were visualized with nitroblue tetrazo-lium (NBT)/5-bromo-4-chloro-3-indolyl phosphate (BCIP).

Measurement of Skin OxygenationSkin oxygenation was measured intradermally in 13 SSc patients and5 healthy controls at the dorsal aspect of the forearm (midwaybetween wrist and elbow) using a PO2 histograph (Eppendorf,Germany). Healthy controls were age- and sex-matched with the SScpatients.

Cell Culture and Hypoxic InductionFibroblast cultures were obtained from skin biopsies of affected skinof additional patients. Control fibroblast cultures (n�5) were ob-tained from healthy subjects. For exposure to hypoxia, fibroblastswere grown to 50% to 80% confluence and transferred into a hypoxicincubator (Forma Scientific) containing 1% O2 v/v (hypoxia, oxygentension 7 mm Hg) or 20% O2 v/v (normoxia, oxygen tension140 mm Hg).

Western Blot Analysis and ImmunofluorescenceCultured cells were removed from the hypoxic incubator and rinsedquickly with ice-cold PBS. After cell lysis, extraction of nuclearproteins, and transfer on nitrocellulose membranes, HIF-1� proteinwas detected using monoclonal mouse anti–HIF-1� mgc3 antibod-ies.10 Similarly, for immunofluorescence on cultured cells, monoclo-nal mouse anti–HIF-1� mgc3 antibodies were used. Bound antibod-ies were visualized using FITC conjugated goat anti-mouseantibodies.

Real-Time PCRIn all stimulation experiments, total RNA was isolated using theTRIzol LS reagent (Gibco/Life Technologies) according to themanufacturer’s protocol. Reverse transcription and TaqMan real-time PCR specific for VEGF-A was performed as described recent-ly.11 18S was used as an endogenous control. Results were quantifiedwith the threshold cycle (CT) and the comparative CT method. Allexperiments as well as all real-time PCR measurements wereperformed in duplicates.

ImmunohistochemistryHIF-1� protein was detected as described recently using monoclonalmouse anti–HIF-1� antibodies (Novus).12 For Flk-1/VEGFR2,monoclonal mouse antibodies were used. Flt-1/VEGFR1 protein wasdetected using polyclonal rabbit antibodies (both from Santa CruzBiotechnology). For double labeling experiments, monoclonalmouse anti-CD68-antibodies (clone PG-M1, Dako) were used. Serialsections of in situ hybridization and VEGFR immunohistochemistrywere stained with polyclonal rabbit anti-von Willebrand-Factorantibodies (A 0082, Dako).

Stimulation With Platelet-Derived Growth Factorand Interleukin-1�Cultured dermal fibroblasts were grown to confluence in 24-wellplates and stimulated with recombinant platelet-derived growthfactor-BB (PDGF-BB) at 10 and 40 ng/mL and with recombinantinterleukin-1� (IL-1�) at 1 pg/mL, 10 pg/mL, and 100 pg/mL.Recombinant proteins were purchased from R&D Systems (Abing-don, UK). Costimulation experiments were performed with 40ng/mL PDGF-BB and 100 pg/mL IL-1�.

StatisticsData are shown as median with range. The Mann-Whitney test wasused for statistical analysis. A value of P�0.05 was consideredstatistical significant.

ResultsVEGF mRNA Is Upregulated in Skin BiopsiesFrom Patients With SScConsistent with recent findings,13 a constitutive expression ofVEGF could be detected in the epidermis of healthy controls(Figure 1A). Whereas in some healthy controls the expressionof VEGF was limited to suprabasal layers (Figure 1A), othersshowed a more scattered pattern of expression throughout theepidermis. As assessed by scoring in randomly chosenhigh-power fields, the mean percentage of keratinocytesexpressing VEGF was 14% (range 0% to 30%). None of thehealthy controls showed an expression of VEGF in dermalcells.

In contrast and despite the lack of sufficient angiogenesisin the SSc skin, an upregulation of VEGF was found inaffected skin biopsies from patients with SSc (Figure 1B).The mean percentage of keratinocytes expressing VEGF wasincreased to 50% (range 0% to 100%, P�0.05 compared withhealthy controls). In addition to the enhanced expression inthe epidermis, an enhanced expression of VEGF could bedetected in the dermis of 13/15 patients. VEGF was expressedby a variety of cell types including fibroblasts and endothelialcells and inflammatory cells (Figure 1C and 1D). For doublelabeling experiments, see online data supplement. Interest-ingly, biopsy specimens taken from clinically noninvolvedskin from the same patients (n�7) showed the same expres-sion pattern of VEGF in each patient as observed for involvedskin biopsies.

Hypoxia Is Present in Skin Lesions of SSc PatientsGiven the strong upregulation of VEGF in skin biopsies ofSSc patients, we next searched for possible stimulators ofVEGF in the SSc skin. Using a PO2 histograph to measuretissue oxygen levels, SSc patients with nonfibrotic skin at thesite of measurement (forearm) did not show different levelsof intracutaneous PO2 values when compared with healthy

110 Circulation Research July 9, 2004



controls. SSc patients with involved (fibrotic) skin at the siteof measurement, however, had strikingly lower levels of PO2

than healthy controls and SSc patients without fibrotic

changes at the forearm (P�0.05; Table). The lower levels ofoxygen in the involved skin were not caused by systemicparameters, because no differences between the groups werefound for arterial oxygen saturation, hemoglobin content,blood pressure, and heart rate (data not shown).

Characterization of the HIF-1� Induction byHypoxia in Dermal FibroblastsBecause the transcription factor HIF-1 is the key moleculeregulating molecular responses to hypoxia, we next aimed tocharacterize the expression of its oxygen-dependent�-subunit (HIF-1�) in cultured dermal fibroblasts underhypoxic conditions. Note that the oxygen concentration usedin these experiments (1% O2) is equivalent to a PO2 value of7 mm Hg, which is close to the 10% percentile measured infibrotic skin areas of SSc patients. An expression of HIF-1�

was found by Western blotting after 6 hours of hypoxicexposure, whereas no signal could be detected after shorterexposure periods as well as in normoxic fibroblasts (Figure2A). The accumulation of HIF-1� protein was maintainedduring prolonged exposure to hypoxia for 12, 24, and 48hours at a similar expression level (Figure 2B). Interestingly,when these experiments were performed under low-serumconditions, the accumulation of HIF-1� under hypoxic con-ditions was considerably lower than in conditions using 10%fetal calf serum. This effect was seen in both SSc and normalfibroblasts (Figure 2B).

The HIF-1� protein needs to be translocated from thecytosol into the nucleus to induce molecular responses tohypoxia via binding to its HIF-binding site present in theflanking regions of HIF-target genes.14 As analyzed byindirect immunofluorescence, HIF-1� was almost completelylocated in the nucleus in dermal fibroblasts as early as 6 hoursafter hypoxic exposure (Figure 2C). No expression of HIF-1�was found by immunofluorescence in cells cultured undernormoxic conditions. Taken together, these data indicate thatHIF-1� can readily be activated under the given parametersin cells derived from SSc patients.

Effects of Hypoxia on VEGF Levels andExpression of HIF-1� Protein in SSc Skin BiopsiesWe analyzed next whether the observed upregulation ofHIF-1� by hypoxia is accompanied by an induction of VEGFin cultured dermal fibroblasts from SSc patients. As expected,hypoxic exposure for 24, 48, and 96 hours led to increasedlevels of VEGF mRNA in all cultures compared with fibro-blasts kept under normoxic conditions (Figure 3). Thisinduction of VEGF mRNA was of statistical significance atall time points (P�0.05) in both SSc and normal fibroblasts.

The experiments mentioned earlier suggested strongly thatthe abundant expression of VEGF in the SSc skin is mediatedby HIF-1� that per se might have been activated by low PO2

levels in fibrotic skin areas of SSc patients. To address thishypothesis, we analyzed the expression of HIF-1� protein byimmunohistochemistry in skin specimens from SSc patientsand healthy controls and compared them with the expressionof VEGF in the same subjects.

Figure 1. A, In situ hybridization for VEGF in a skin section froma healthy subject showing a constitutive expression of VEGF insuprabasal keratinocytes (NBT/BCIP development, dark bluesignals), but no expression in the dermis. B, In situ hybridizationfor VEGF in an affected skin section from a patient with SScshowing a strong upregulation of VEGF mRNA in the epidermisas well as in dermal structures. Magnifications �100. C, Highermagnification of a skin section from an additional SSc patientshowing expression of VEGF mRNA in endothelial cells (arrows)and (D) dermal fibroblasts.

Distler et al VEGF and HIF-1� in Systemic Sclerosis 111



In skin specimens from healthy controls, nuclear signals forHIF-1� were detected in the epidermis with a moderate to highexpression in all subjects (Figure 4A). (For detailed semiquan-titative analysis, see online Table 2.) VEGF showed a similarpattern of expression in serial sections from the same healthy

controls (Figure 4C). This coexpression of HIF-1� and VEGF inhealthy subjects indicates that the constitutive expression ofVEGF in the epidermis might in fact be driven by HIF-1�. Inparallel with the results for VEGF, no expression of HIF-1�could be detected in the dermis of healthy controls.

Intracutaneous Levels of PO2 From Healthy Controls and SSc Patients

Mean PO2, mm Hg Median PO2, mm Hg 10% Percentile PO2, mm Hg

Healthy controls 33.6�4.1 32.9�4.2 21.1�2.7

SSc nonfibrotic 37.9�8.6 38.3�9.8 22.3�5.6

SSc fibrotic 23.7�2.1* 21.7�4.0* 8.9�3.2*

Levels of PO2 were measured at the forearm by 400 single PO2 measurements in each subject. Mean,median, and 10% percentile PO2 values were calculated for each subject. SSc patients with fibrotic skinat the site of measurements showed significantly lower levels of PO2 than healthy controls and SSc patientswithout fibrotic changes at the site of measurement in all parameters analyzed. *P�0.05.

Figure 2. Representative Western blots forHIF-1� using nuclear protein extracts fromcultured cells. HELA cells were used as posi-tive controls. Anti–Sp-1 antibodies were usedto control for equal loading of proteins A,Time-dependent induction of HIF-1� in SScfibroblasts under hypoxic conditions. Expres-sion of HIF-1� could be detected after 6hours, whereas no signal was found aftershorter exposure times to hypoxia and in nor-moxic controls. B, Effects of long-term hypox-ia (48 hours) and of low-growth factor condi-tions on HIF-1� expression. Expression ofHIF-1� in dermal fibroblasts was maintainedafter prolonged exposure to hypoxia at a sim-ilar level of expression. Low growth factorconditions (1% FCS) reduced the expressionof HIF-1�. C, Immunofluorescence with anti–HIF-1� antibodies on SSc dermal fibroblastsafter 6 hours of hypoxic incubation (1% O2)showing almost complete translocation intothe nucleus.




Surprisingly, the expression of HIF-1� protein in skinspecimens from SSc patients was lower than in those fromhealthy subjects. In the epidermis, we found weak signals forHIF-1� in a limited number of keratinocytes, which con-trasted with the abundant epidermal expression of VEGF inthese biopsies (Figure 4B and 4D). In addition, the expressionpattern of HIF-1� did not correlate with the expressionpattern of VEGF. Consistent with this observation, noHIF-1� protein was detected in dermal cells of SSc patients,despite the strong upregulation of VEGF in the dermis. Takentogether, these data suggest that HIF-1–independent mecha-nisms contribute to the upregulation of VEGF in the SSc skin.

Expression of VEGF in Dermal FibroblastsTo compare the normoxic constitutive synthesis of VEGF indermal fibroblasts from SSc patients and healthy controls,cells were grown to confluence and quantified for VEGFexpression by real-time PCR. VEGF mRNA could be de-tected in all healthy and SSc fibroblast cultures with somevariability between individual fibroblasts in both groups. Incontrast to the observation in skin biopsies, no significantdifference were detected between SSc and normal fibroblasts

(median �Ct normal: 8.2, range 6.2 to 8.6; median �Ct SSc:8.4, range 6.4 to 8.7). Similarly, under low growth factorconditions (1% FCS), no differences were found betweenfibroblasts derived from healthy controls and SSc patients(median �Ct normal: 18.0, range 17.2 to 18.7; median �CtSSc: 17.8, range 17.3 to 19.1).

Cytokines Induce VEGF in Dermal FibroblastsBecause HIF-1� does not appear to be a major mediator ofVEGF upregulation in the SSc skin, we focused on PDGF aswell as IL-1, both factors being implicated as key moleculesin the pathogenesis of SSc in a number of previous stud-ies.15–18 When fibroblasts were treated with recombinantPDGF, there was a dose-dependent increase of VEGF mRNAin both SSc and normal fibroblasts (10 ng/mL: normal,1.6�0.4-fold increase; SSc, 1.6�0.3-fold increase; 40 ng/mL: normal, 3.0�0.8 fold increase; SSc, 1.8�0.5-fold in-crease; P�0.05 compared with nonstimulated controls).

Treatment with recombinant IL-1� resulted also in adose-dependent increase of VEGF mRNA in all fibroblastcultures. Inductions were small but significant, reaching a2.4�0.6-fold increase of VEGF mRNA in SSc fibroblasts

Figure 3. Synthesis of VEGF mRNA in dermalfibroblasts after exposure to hypoxia. Mean andSD fold increase of VEGF mRNA comparedwith normoxic controls (basal levels, defined as1). A statistically significant increase of VEGFwas found in normal as well as SSc fibroblastsat all time points. #P�0.05.

Figure 4. A and B, Immunohistochemis-try for HIF-1� in skin specimens fromhealthy controls (A) and patients withSSc (B). Healthy controls showed anuclear expression of HIF-1� in keratino-cytes (red signal), whereas the expres-sion of HIF-1� was weak or absent inSSc patients. Development with FastRed chromogen. C and D, In situ hybrid-ization for VEGF in skin sections fromthe same controls (C) and SSc patients(D). Whereas HIF-1� coexpressed to acertain extent with VEGF in the epider-mis of healthy controls, the weak orabsent expression of HIF-1� in SScpatients contrasted with the abundantexpression of VEGF in the samepatients.




and 1.9�0.2-fold increase in normal fibroblasts at concentra-tions of 100 pg/mL (P�0.05 compared with nonstimulatedcontrols).

In vivo, SSc dermal fibroblasts are exposed to a variety ofcytokines that are overexpressed in the SSc skin includingPDGF and IL-1�.15–18 To mimic this situation, cultured cellswere costimulated with recombinant PDGF and IL-1� andanalyzed for VEGF mRNA levels. Whereas the stimulationwith either PDGF or IL-1� resulted in a rather small increaseof VEGF, the costimulation with 100 pg/mL IL-1� and 40ng/mL PDGF showed additive effects with a strong increaseof VEGF (Figure 5). The induction of VEGF raised to4.7�0.5-fold increase in SSc fibroblasts compared with a2.0�0.4 fold increase with IL-1� alone and a 2.3�0.2 foldincrease with PDGF alone. These data indicate that the strongupregulation of VEGF in the SSc skin might be a net effect ofdifferent cytokines rather than an effect of a single factor.

Upregulation of VEGF Receptors inSystemic SclerosisPossible explanations for the lack of a sufficient angiogenesisin the SSc skin despite the upregulation of VEGF include areduced expression of VEGF receptors (VEGFRs) on endo-thelial cells. To address this possibility, immunohistochem-istry with antibodies against Flt-1/VEGFR1 and Flk-1/VEGFR2 was performed on skin sections of SSc patients andhealthy controls. Expression of Flt-1/VEGFR1 was found inall biopsies from SSc patients (Figure 6A). VEGFR1 has beendescribed as an endothelial specific molecule.19 Consistentwith these reports, the expression of VEGFR1 was limited toendothelial cells of smaller vessels in the SSc skin (see alsoonline Figure 3). In contrast to the significant expression inthe SSc skin, expression of VEGFR1 could not be detected in5/6 biopsies from healthy controls (Figure 6B), whereas onehealthy control showed a weak expression on endothelialcells.

Similar to VEGFR1, the expression of Flk-1/VEGFR2 wasfound to be upregulated in SSc skin biopsies (Figure 6C). Inagreement with the in situ hybridization for VEGF, VEGFR2was found expressed in 13/14 patients. Again, the expressionof VEGFR2 was preferentially detected on endothelial cells(see also online Figure 4). In general, the number of endo-thelial cells expressing VEGFR2 and the intensity of stainingwas higher than for VEGFR1. No other dermal structuresthan endothelial cells expressed VEGFR2, whereas somepositive signals were observed in the epidermis from bothSSc patients and healthy controls (Figure 6C and 6D).Interestingly, whereas in contrast to normal skin, both VEGFreceptors were expressed also in biopsies from noninvolved(nonfibrotic) skin of SSc patients, the percentage of endothe-lial cells expressing both receptors was lower than in biopsiesof involved (fibrotic) skin of the same patients (see onlineTables 3 and 4 for detailed semiquantitative analysis ofVEGFR1 and VEGFR2 immunohistochemistry).

Figure 5. Mean and SD fold induction of VEGF after stimulationwith PDGF and IL-1� alone as well as after costimulation withPDGF and IL-1� compared with nonstimulated controls (basallevels, defined as 1). #P�0.05 compared with stimulation witheach cytokine alone; *P�0.05 compared with nonstimulatedcontrols.

Figure 6. A and B, Immunohistochemis-try for flt-1/VEGFR1 on skin specimensfrom patients with SSc (A) and healthycontrols (B). Expression of flt-1/VEGFR1was found on endothelial cells in skinsections from SSc patients (brownish-redsignal), whereas no expression could bedetected in healthy controls. Develop-ment with AEC chromogen. C and D,Immunohistochemistry for flk-1/VEGFR2on skin specimens from patients withSSc (C) and healthy controls (D). Similarto VEGFR1, a strong expression of flk-1/VEGFR2 could be detected on endotheli-al cells of patients with SSc (dark bluesignal), whereas no expression wasfound in healthy controls. Developmentwith NBT/BCIP.




DiscussionIn the present study, we addressed the hypothesis that the lackof sufficient angiogenesis in SSc might be explained by adownregulation of the potent angiogenic factor VEGF, whichis a key molecule in several steps of angiogenesis.20 How-ever, unexpectedly, skin specimens of patients with SScshowed a strong upregulation of VEGF by in situ hybridiza-tion compared with healthy controls. Apart from VEGF, alsoits receptors VEGF-R1 and VEGF-R were found to beupregulated on endothelial cells of SSc patients in vivo. Inaddition, we could show recently that VEGF protein wassignificantly increased in blood samples from patients withSSc,21 reaching levels observed in patients with numerousmalignant diseases.22,23 Taken together, these data stronglyimply an activation of the VEGF/VEGF-receptor axis inpatients with SSc.

Angiogenesis is regulated by a tightly controlled balance ofangiogenic and angiostatic factors.5 Thus, an attractive hy-pothesis for the insufficient angiogenesis in SSc despiteupregulated increased levels of VEGF could be an upregula-tion of angiostatic factors that is even higher than those ofVEGF and thereby outweighs its proangiogenic effects.However, data on angiostatic factors in SSc are inconsistent.For instance, the collagen type XVIII breakdown productendostatin has been described as significantly increased ornot different from healthy controls.21,24 Recently, Macko etal25 found elevated levels of platelet factor-4 andthrombospondin-1 in plasma samples from patients with SSc.

Of particular interest regarding the role of VEGF in thepathogenesis of SSc is the increasing experimental evidenceabout the diverse biological actions of VEGF depending onthe period and level of expression.26 Using pTET-VEGF165/MHC�-tTa transgenic mice, in which the expression ofVEGF can be conditionally switched off in an organ-dependent manner by feeding tetracycline, Dor et al27 showedthat overexpression of VEGF induces the formation of newfunctional vessels in adult organs. These vessels were matureas reflected by the presence of smooth muscle cells andresulted in an improved organ perfusion. However, mostinterestingly, prolonged exposure to VEGF without subse-quently switching off its expression resulted in the formationof irregularly shaped sac-like vessels with reduced bloodflow, very much reminiscent of the disturbed vessel morphol-ogy with megacapillaries seen in SSc patients. In fact, thetimely downregulation of VEGF rather then the induction ofangiostatic factors appears to be the major mechanism pre-venting the formation of chaotic vessel morphology. This isfurther evidenced by the association of chronic overexpres-sion of VEGF with glomeruloid and hemangioma-like vesselsin other experimental settings.28,29 Thus, although short-timeupregulation of VEGF is a strong inducer of angiogenesis,chronic and uncontrolled overexpression of VEGF leads tochaotic vessel morphology with reduced blood flow in thenewly formed vessels.

In the present study, overexpression of VEGF in the skinspecimens was detected independent of the disease durationin both patients with early as well as late disease. In addition,recent detailed analysis of blood levels in SSc revealed thatVEGF was significantly upregulated in patients with different

disease durations including patients with very early stagesthat not yet fulfilled ACR criteria, but did so during a 1-yearfollowup.21 Furthermore, cytokines known to induce thesynthesis of VEGF are overexpressed in SSc throughout thedisease course. In addition to PDGF and IL-1, this alsoincludes TGF�, which is considered a key factor in thedevelopment of fibrosis in SSc.30,31 Taken together, these dataindicate that a chronic and uncontrolled overexpression ofVEGF does occur in SSc and might significantly contribute tothe chaotic capillary morphology seen in these patients. Theuncontrolled overexpression of VEGF does already exist inearliest disease stages in parallel with the morphologicalcapillary changes observed by capillary microscopy. Thechronic expression appears to be driven by key cytokines inthe pathogenesis of SSc including PDGF and IL-1 throughoutdifferent disease stages.

We also addressed other possible stimulators of VEGF inthe SSc skin. Surprisingly, despite the clear presence ofhypoxia in the involved skin of patients with SSc and despitethe nuclear accumulation of HIF-1� in cells derived from SScpatients under hypoxic conditions in vitro, the expression ofHIF-1� protein was lower in skin specimens from SScpatients than in healthy controls. Although in vitro thestabilization of HIF-1� protein is instantaneous32 and sus-tained for up to 48 hours in dermal fibroblasts, the responsein vivo might be substantially different. Consistent with ourfindings, it has been reported that exposure of mice to 6%hypoxia leads to maximum levels of HIF-1� protein in thebrain after 4 to 5 hours, but declines afterward, reaching basalnormoxic levels after 9 to 12 hours.33 Similar results wereobtained for the kidney and the liver. Considering thecorrelation of low oxygen levels with fibrotic structuralchanges in the skin of patients with SSc in the present study,hypoxia in SSc patients appears to be chronic rather thancaused by acute situations such as vasoconstriction in thesetting of Raynaud’s phenomenon. Whether the response ofHIF-1� to chronic hypoxia is similar in the skin as reportedfor other organs has not yet been addressed. It has to bestressed, however, that our findings do not rule out thatpathways induced by hypoxia independent from HIF-1� (eg,via induction of other members of the HIF family) signifi-cantly contribute to the pathogenesis of SSc.34 This is furthersupported by recent data showing that hypoxia directlycontributes to the accumulation of extracellular matrix bothby HIF-1�– dependent as well as HIF-1�–independentmechanisms.35

In summary, we have shown that, despite the lack of asufficient angiogenesis in SSc, the proangiogenic factorVEGF together with its receptors VEGF-R1 and VEGF-R areupregulated in skin specimens from SSc patients. Althoughinvolved skin is characterized by strongly reduced levels ofoxygen, hypoxia-induced HIF-1� does not appear to be themajor mediator of VEGF induction in SSc. Instead, thechronic and uncontrolled overexpression of VEGF is basedon a net effect of cytokines such as IL-1 and PDGF. SScmight therefore serve as a disease model for the effects of achronic and unterminated upregulation of VEGF with theformation of irregularly shaped sac-like megacapillaries.Consequently, for therapeutic approaches aiming to improve




tissue perfusion in these patients, a controlled expression andtimely termination of VEGF signaling appears to be crucialfor success of proangiogenic therapies.

AcknowledgmentsThe study was supported by the Deutsche Akademie der Naturfor-scher Leopoldina (grant BMBF-LPD 9901/8-7), the Braun Founda-tion, and grant 560031 of the University of Zurich. The authors thankFerenc Pataky and Maria Comazzi for excellent technical assistance.

References1. Furst DE, Clements PJ. Pathogenesis, Fusion (Summary). In: Furst DE,

Clements PJ, eds. Systemic Sclerosis. Baltimore, Md: Williams &Wilkins;1996:275–284.

2. Cutolo M, Grassi W, Matucci Cerinic M. Raynaud’s phenomenon and therole of capillaroscopy. Arthritis Rheum. 2003;48:3023–3030.

3. McHugh NJ, Distler O, Giacomelli R, Riemekasten G. Non organ basedlaboratory markers in systemic sclerosis. Clin Exp Rheumatol. 2003;21:S32–S38.

4. LeRoy EC. Systemic sclerosis: a vascular perspective. Rheum Dis ClinNorth Am. 1996;22:675–694.

5. Distler J, Hirth A, Kurowska-Stolarska M, Gay RE, Gay S, Distler O.Angiogenic and angiostatic factors in the molecular control of angio-genesis. Q J Nucl Med. 2003;47:149–161.

6. Hofer T, Wenger H, Gassmann M. Oxygen sensing, HIF-1alpha stabili-zation and potential therapeutic strategies. Pflugers Arch. 2002;443:503–507.

7. Preliminary criteria for the classification of systemic sclerosis(scleroderma). Subcommittee for scleroderma criteria of the Am Rheu-matism Association Diagnostic and Therapeutic Criteria Committee.Arthritis Rheum 1980;23:581–590.

8. Plate KH, Breier G, Weich HA, Risau W. Vascular endothelial growthfactor is a potential tumour angiogenesis factor in human gliomas in vivo.Nature. 1992;359:845–848.

9. Kowal-Bielecka O, Distler O, Neidhart M, Kunzler P, Rethage J, NawrathM, Carossino A, Pap T, Muller-Ladner U, Michel BA, Sierakowski S,Matucci-Cerinic M, Gay RE, Gay S. Evidence of 5-lipoxygenase over-expression in the skin of patients with systemic sclerosis: a newly iden-tified pathway to skin inflammation in systemic sclerosis. ArthritisRheum. 2001;44:1865–1875.

10. Camenisch G, Tini M, Chilov D, Kvietikova I, Srinivas V, Caro J,Spielmann P, Wenger RH, Gassmann M. General applicability of chickenegg yolk antibodies: the performance of IgY immunoglobulins raisedagainst the hypoxia-inducible factor 1alpha. FASEB J. 1999;13:81–88.

11. Distler JH, Hagen C, Hirth A, Muller-Ladner U, Lorenz HM, Del RossoA, Michel BA, Gay RE, Nanagara R, Nishioka K, Matucci-Cerinic M,Kalden JR, Gay S, Distler O. Bucillamine induces the synthesis ofvascular endothelial growth factor dose-dependently in systemic sclerosisfibroblasts via nuclear factor-�B and simian virus 40 promoter factor 1pathways. Mol Pharmacol. 2004;65:389–399.

12. Krieg M, Haas R, Brauch H, Acker T, Flamme I, Plate KH. Up-regulationof hypoxia-inducible factors HIF-1� and HIF-2� under normoxic con-ditions in renal carcinoma cells by von Hippel-Lindau tumor suppressorgene loss of function. Oncogene. 2000;19:5435–5443.

13. Brown LF, Harrist TJ, Yeo KT, Stahle-Backdahl M, Jackman RW, BerseB, Tognazzi K, Dvorak HF, Detmar M. Increased expression of vascularpermeability factor (vascular endothelial growth factor) in bullous pem-phigoid, dermatitis herpetiformis, and erythema multiforme. J InvestDermatol. 1995;104:744–749.

14. Wenger RH. Cellular adaptation to hypoxia: O2-sensing proteinhydroxylases, hypoxia-inducible transcription factors, and O2-regulatedgene expression. FASEB J. 2002;16:1151–1162.

15. Aotsuka S, Nakamura K, Nakano T, Kawakami M, Goto M, Okawa-Takatsuji M, Kinoshita M, Yokohari R. Production of intracellular andextracellular interleukin-1 alpha and interleukin-1 beta by peripheralblood monocytes from patients with connective tissue diseases. AnnRheum Dis. 1991;50:27–31.

16. Gay S, Jones RE, Jr., Huang GQ, Gay RE. Immunohistologic demon-stration of platelet-derived growth factor (PDGF) and sis-oncogeneexpression in scleroderma. J Invest Dermatol. 1989;92:301–303.

17. Ludwicka A, Ohba T, Trojanowska M, Yamakage A, Strange C, SmithEA, LeRoy EC, Sutherland S, Silver RM. Elevated levels of plateletderived growth factor and transforming growth factor-beta 1 in bron-choalveolar lavage fluid from patients with scleroderma. J Rheumatol.1995;22:1876–1883.

18. Umehara H, Kumagai S, Murakami M, Suginoshita T, Tanaka K, HashidaS, Ishikawa E, Imura H. Enhanced production of interleukin-1 and tumornecrosis factor alpha by cultured peripheral blood monocytes frompatients with scleroderma. Arthritis Rheum. 1990;33:893–897.

19. Neufeld G, Cohen T, Gengrinovitch S, Poltorak Z. Vascular endothelialgrowth factor (VEGF) and its receptors. FASEB J. 1999;13:9–22.

20. Mackiewicz Z, Sukura A, Povilenaite D, Ceponis A, Virtanen I,Hukkanen M, Konttinen YT. Increased but imbalanced expression ofVEGF and its receptors has no positive effect on angiogenesis in systemicsclerosis skin. Clin Exp Rheumatol. 2002;20:641–646.

21. Distler O, Del Rosso A, Giacomelli R, Cipriani P, Conforti ML, GuiducciS, Gay RE, Michel BA, Bruhlmann P, Muller-Ladner U, Gay S, Matucci-Cerinic M. Angiogenic and angiostatic factors in systemic sclerosis:increased levels of vascular endothelial growth factor are a feature of theearliest disease stages and are associated with the absence of fingertipulcers. Arthritis Res. 2002;4:R11.

22. Iwasaki T, Hamano T, Ogata A, Hashimoto N, Kitano M, Kakishita E.Clinical significance of vascular endothelial growth factor and hepatocytegrowth factor in multiple myeloma. Br J Haematol. 2002;116:796–802.

23. Poon RT, Lau CP, Cheung ST, Yu WC, Fan ST. Quantitative correlationof serum levels and tumor expression of vascular endothelial growthfactor in patients with hepatocellular carcinoma. Cancer Res. 2003;63:3121–3126.

24. Hebbar M, Peyrat JP, Hornez L, Hatron PY, Hachulla E, Devulder B.Increased concentrations of the circulating angiogenesis inhibitorendostatin in patients with systemic sclerosis. Arthritis Rheum. 2000;43:889–893.

25. Macko RF, Gelber AC, Young BA, Lowitt MH, White B, Wigley FM,Goldblum SE. Increased circulating concentrations of the counterad-hesive proteins SPARC and thrombospondin-1 in systemic sclerosis(scleroderma). Relationship to platelet and endothelial cell activation.J Rheumatol. 2002;29:2565–2570.

26. Carmeliet P. VEGF gene therapy: stimulating angiogenesis or angioma-genesis? Nat Med. 2000;6:1102–1103.

27. Dor Y, Djonov V, Abramovitch R, Itin A, Fishman GI, Carmeliet P,Goelman G, Keshet E. Conditional switching of VEGF provides newinsights into adult neovascularization and pro-angiogenic therapy. EMBOJ. 2002;21:1939–1947.

28. Drake CJ, Little CD. Exogenous vascular endothelial growth factorinduces malformed and hyperfused vessels during embryonic neovascu-larization. Proc Natl Acad Sci U S A. 1995;92:7657–7661.

29. Sundberg C, Nagy JA, Brown LF, Feng D, Eckelhoefer IA, Manseau EJ,Dvorak AM, Dvorak HF. Glomeruloid microvascular proliferationfollows adenoviral vascular permeability factor/vascular endothelialgrowth factor-164 gene delivery. Am J Pathol. 2001;158:1145–1160.

30. Kissin EY, Korn JH. Fibrosis in scleroderma. Rheum Dis Clin North Am.2003;29:351–369.

31. Pertovaara L, Kaipainen A, Mustonen T, Orpana A, Ferrara N, Saksela O,Alitalo K. Vascular endothelial growth factor is induced in response totransforming growth factor-beta in fibroblastic and epithelial cells. J BiolChem. 1994;269:6271–6274.

32. Jewell UR, Kvietikova I, Scheid A, Bauer C, Wenger RH, Gassmann M.Induction of HIF-1alpha in response to hypoxia is instantaneous. FASEBJ. 2001;15:1312–1314.

33. Stroka DM, Burkhardt T, Desbaillets I, Wenger RH, Neil DA, Bauer C,Gassmann M, Candinas D. HIF-1 is expressed in normoxic tissue anddisplays an organ-specific regulation under systemic hypoxia. FASEB J.2001;15:2445–2453.

34. Hofer T, Wenger RH, Kramer MF, Ferreira GC, Gassmann M. Hypoxicup-regulation of erythroid 5-aminolevulinate synthase. Blood. 2003;101:348–350.

35. Distler J, Hirth A, Scheid A, Del Rosso A, Gay RE, Lorenz HM, KaldenJR, Matucci-Cerinic M, Gassmann M, Gay S, Distler O. Expressionprofile of SSc fibroblasts reveals extracellular matrix proteins as down-stream molecules of hypoxia induced pathways. Arthritis Rheum. 2002;46:S203. Abstract.




2

Materials and Methods

Patients and skin biopsies

Skin biopsies were obtained from 15 patients who met the American College of

Rheumatology criteria for SSc 1. In all patients, biopsies were taken from clinically involved

skin. In a subset of patients (n = 7), biopsies were also taken from clinically non-involved skin

as assessed by skin scoring of an experienced examiner. Tissue samples of 6/15 patients were

fixed in 4% buffered formalin for 6 hours, transferred to 50% ethanol and embedded in

paraffin 2. Skin biopsies of 9/15 patients with SSc were snap frozen in O.C.T. Tissue Tek

embedding medium (Miles, Elkhart, IN) and stored at -80° C. Controls (n = 6) consisted of

biopsies from healthy volunteers. Experiments were approved by local ethics review

committees, and written informed consent was obtained from all patients.

Cell culture and hypoxic induction

Fibroblast cultures were obtained from skin biopsies of affected skin of additional patients.

Control fibroblast cultures (n = 5) were obtained from healthy subjects. After enzymatic

digestion of the skin biopsies with Dispase II (Boehringer-Mannheim, Rotkreuz, Switzerland),

fibroblasts were grown in DMEM containing 10% heat inactivated FCS. Fibroblasts from

passages 3-9 were used for the experiments.

For exposure to hypoxia, fibroblasts were grown to 50-80% confluence and medium was

changed 24 hours before initiation of the experiments. Fibroblasts were transferred into a

hypoxic incubator (Forma Scientific, Illkirch, France) and exposed to a humidified

atmosphere containing 5% CO2 and 1% O2 v/v (hypoxia, oxygen tension 7 mmHg) at 37°C

for 6 to 96 hours 3. For control, cells were cultured under the same conditions except that the

atmosphere contained 20% O2 v/v (normoxia, oxygen tension 140 mmHg). In experiments



3

under low growth factor conditions, medium was changed from DMEM/10% FCS to

DMEM/1% FCS 24 hours before the experiments.

Measurement of skin oxygenation

Skin oxygenation was measured intradermally in 13 SSc patients and 5 healthy controls at the

dorsal aspect of the forearm (midway between wrist and elbow) using a pO2 histograph

(Eppendorf, Germany). The healthy controls were age- and sex-matched with the SSc

patients. To exclude circadian and seasonal influences on the results, all measurements were

taken in the morning between 9 and 11 a.m. and during the same season (January-March).

The pO2 histograph consists of a microprocessor-guided polarographic electrode system and

is considered the gold standard for the measurement of tissue oxygen tensions 4;5. Patients and

controls were allowed to adapt to room temperature for at least 30 minutes. All measurements

were performed with the subjects sitting in an upright position their forearm rested at a 90°

angle. After local anesthesia with 1% Lidocaine (Sintetica, Switzerland), the skin was

punctured with a needle, and the polarographic needle electrode was inserted 3-4 mm into the

tissue. The electrode was then moved automatically through the tissue in steps of 0.4 mm with

a forward movement of 0.7 mm followed immediately by a backward movement of 0.3 mm to

minimize pressure artifacts. After every step, the pO2 value was recorded. When 50

measurements were taken, the needle electrode was removed and the procedure was repeated

in a different direction thereby recording 400 pO2 values in a 4-cm2 area of the middle

forearm in each subject. To ensure that the pO2 values were measured in the dermis, the

location of the needle electrode was controlled by ultrasound. At the end of the experiment,

blood gas analysis was performed by puncture of the A. radialis. In addition, blood pressure

and hemoglobin were determined by standardized procedures. The mean duration of the

experiment was 70 ± 10 minutes without differences between the groups. Initial experiments



4

with healthy volunteers and measurements at different days showed that the results were

reproducible and were not influenced by the local anesthesia used.

Western blot analysis and immunofluorescence

Cultured cells were removed from the hypoxic incubator and rinsed quickly with ice-cold

PBS. Cell-lysis buffer consisting of 10 mM Tris/HCl (pH 8.0), 1 mM EDTA (pH 8.0), 150

mM NaCl, 0.1% Nonidet P-40, 1 mM phenylmethylsulphonylfluoride (PMSF), 1 mM

Na3VO4 and aprotinin/leupeptin/pepstatin A (all 1 µg/ml) was then immediately applied.

After incubation on ice for 10 minutes, the homogenate was centrifuged at 1600 g and 4°C for

5 minutes. Nuclear extraction buffer containing 20 mM hydroxyethyl-piperazine-

ethanesulfonic acid (HEPES, pH 7.5), 400 mM NaCl, 1 mM EDTA (pH 8.0), 1 mM

dithiothreitol (DTT) and 1 mM PMSF was added to the pellet in a 1:1 ratio and incubated on

ice for 15 minutes. After centrifugation (20 000g, 5 minutes, 4°C), supernatants were

collected and stored at -80°C.

The Bradford assay (Biorad, Reinach, Switzerland) was used to determine protein

concentrations. Thirty µg of nuclear extract per sample were separated by 7.5% SDS-PAGE

and electrotransferred onto nitrocellulose membranes according to standard protocols 6.

Protein transfer and equal loading were confirmed by Ponceau S staining. After blocking with

4% nonfat milk powder in phosphate-buffered saline (PBS) for 1 h, immunoblots were

incubated with monoclonal mouse anti-HIF-1α mgc3 antibodies 7 for 2 hours at a dilution of

1:500, and with horseradish peroxidase-conjugated goat anti-mouse antibodies

(Promega/Catalys, Wallisellen, Switzerland) for 1 hour at a dilution of 1:10000.

Chemiluminescence detection was performed by incubation with 100 nM Tris/HCl (pH 8.5),

2.65 mM H2O, 0.45 mM luminol and 0.625 mM coumaric acid for 1 min, followed by

exposure to X-ray films (SuperRX, Fuji, Dielsdorf, Switzerland). To verify equal loading of

proteins, immunoblots were incubated with polyclonal rabbit anti-Sp1 antibodies (Santa Cruz,



5

Santa Cruz, CA; dilution 1:1000), followed by horseradish peroxidase-conjugated anti-rabbit

antibodies (Sigma; dilution 1:10000).

For immunofluorescence, cells were cultured to sub-confluence in chamber slides, and

exposed to hypoxia as described. Cells were then fixed immediately in freshly prepared 4%

formaldehyde (pH 7.0) for 10 minutes. After incubation with 0.5% Triton X for 5 minutes and

blocking with 10% FCS in PBS for 30 minutes, cells were incubated for 1 hour at 37°C with

monoclonal mouse anti-HIF-1α mgc3 antibodies (dilution 1:10) or with isotype matched IgG

antibodies (Dako, Zug, Switzerland) for control, followed by FITC conjugated goat anti-

mouse antibodies (Jackson ImmunoResearch/Milan Analytica, LaRoche Switzerland, dilution

1:400) for 30 minutes at room temperature.

Immunohistochemistry

HIF-1α protein was detected by immunohistochemistry on paraffin embedded sections of skin

biopsies from SSc patients and healthy controls as described recently 8. In brief, after

deparaffinization and rehydration, endogenous peroxidase was blocked with 100%

methanol/1%H2O2 at 4°C for 20 minutes. Antigen retrieval was performed using the Dako

antigen retrieval system according to the instructions of the manufacturer. Sections were

rinsed in PBS and incubated with 5% bovine serum albumin (BSA, Sigma) for 30 minutes

and with 20% normal goat serum (NGS, Sigma) for 60 minutes at room temperature to block

non-specific binding. Incubation with monoclonal mouse anti-HIF-1α antibodies (dilution 1:

200, Novus, Littleton, CO, USA) was performed overnight at 4°C in 10% NGS/PBS. After

rinsing in PBS/0.1% Triton and incubation with 20% NGS/PBS/0.1% Triton for 1 hour at

room temperature, sections were exposed to biotinylated rat anti-mouse antibodies (dilution 1:

300 in NGS/PBS/0.1% Triton, Dianova, Hamburg, Germany) for 1 hour. Sections were then

incubated with alkaline phosphatase conjugated streptavidin (Vector, Burlingame, CA, USA)

for 1 hour at room temperature, followed by visualization with Fast Red (Sigma). Control



6

experiments were performed with isotype-matched antibodies instead of the primary

antibodies.

For the detection of Flk-1/VEGFR2, monoclonal mouse antibodies (Santa Cruz

Biotechnology, Santa Cruz, USA) were used. Sections were exposed to microwave heating in

10 nM sodium citrate (pH 6.0) for antigen retrieval. Unspecific binding was blocked by

incubation with 4% milk and 2% normal horse serum in Tris buffer (pH = 7.6) for 30 minutes

at room temperature. Then, slides were incubated with anti-Flk-1/VEGFR-2 antibodies (1:50

dilution in Tris) for 2 hours, with secondary antibodies (goat anti-mouse IgG + IGM, H + L,

diluted 1:400 in Tris; Jackson ImmunoResearch Laboratories, West Grove, USA) for 30

minutes and with APAAP complexes (1:50 dilution in Tris; Dako) for 30 minutes. All steps

were performed at room temperature and with Tris/0.1% saponin as washing solution. Bound

APAAP complexes were visualized with NBT/BCIP in 10% polyvinyl alcohol (molecular

weight 70-100 kD). Negative controls consisted of isotype matched IgG1 antibodies instead of

the primary antibodies.

For Flt-1/VEGFR1 immunohistochemistry, a biotin-streptavidin based amplifying system

(BioGenex, San Ramon, USA) with polyclonal rabbit IgG antibodies against flk-1 (Santa

Cruz Biotechnology) was used. Paraffin embedded sections were processed as described

above. After antigen retrieval, endogenous peroxidase activity was blocked with 3% H2O2 for

5 minutes. Slides were incubated for 1 hour with anti-flt-1 antibodies (1:50 dilution in Tris)

followed by incubation with biotinylated anti-immunglobulins for 20 minutes. Labeling was

performed for 20 minutes using a horseradish peroxidase-conjugated streptavidin complex

with high affinity for biotin (BioGenex). Antigens were visualized using aminoethylcarbazole

(AEC) chromogen and hydrogen peroxide as substrate. For control, antibodies were

preincubated with an excess of blocking peptide (10 times the concentration of the primary

antibodies; Santa Cruz Biotechnology) for 1 hour at room temperature. This solution was then

applied instead of the primary antibodies.



7

The results of each immunohistochemistry were quantified using a score ranging from no

expression to high expression: (+) = weak staining of a few cells, + = <25% of cells stained

positive, ++ = 25-50% positive cells, +++ > 50% positive cells.

For double-labeling experiments after in situ hybridization for VEGF, monoclonal mouse

anti-CD68 antibodies (clone PG-M1, Dako, Glostrup, Denmark) were used. Unspecific

binding was blocked by incubation with 4% milk and 2% normal horse serum in Tris buffer

(pH = 7.6) for 30 minutes. Slides were incubated with anti-CD68 antibodies at a dilution of

1:60 in Tris buffer for 1 hour. Bound primary antibodies were linked with biotinylated anti-

immunglobulins (BioGenex) for 20 minutes. Labeling was performed for 20 minutes using a

horseradish peroxidase-conjugated streptavidin complex with high affinity for biotin

(BioGenex). All steps were performed at room temperature. Antigens were visualized using

aminoethylcarbazole (AEC) chromogen and hydrogen peroxide as substrate. In control

experiments, matched mouse IgG isotypes were used instead of the primary antibodies.

Serial section of in situ hybridization for VEGF as well as of immunohistochemistry for Flk-

1/VEGFR2 and Flt-1/VEGFR1 were stained with polyclonal rabbit anti-von Willebrand-

Factor antibodies (A 0082, Dako). After blocking with 1% H2O2/0.01% NaN3 for 5 minutes,

slides were incubated with the polyclonal rabbit anti-von Willebrand-Factor antibodies at a

dilution of 1:2000 overnight at 4°C. The secondary antibody (biotin-labeled goat-anti-rabbit

antibodies) was applied at a dilution of 1:1000 for 30 minutes at room temperature. Labeling

was performed using the Vectastain ABC kit containing peroxidase (Vector Laboratories,

Burlingame, CA, USA), and antigens were visualized with Nova RED (Vector Laboratories).

In control experiments, irrelevant rabbit IgGs were used instead of the primary antibodies.

Stimulation with PDGF and IL-1β

To examine the effects of platelet-derived growth factor (PDGF) and interleukin-1β (IL-1β),

which are known to be upregulated in the clinically involved skin of patients with SSc,



8

fibroblasts were cultured in 24-well plates. For stimulation with PDGF, fibroblasts were

grown to confluence; DMEM/10%FCS was replaced with 1.5 ml DMEM without FCS and

incubated for 24 hours. Recombinant PDGF-BB (R&D Systems, Abingdon, UK) stored at -

80°C in stock concentration of 10 µg/ml in PBS/0.1% bovine serum albumin/4mM HCl was

used for the experiments. Cells were washed with PBS, and recombinant PDGF-BB dissolved

in 1.5 ml DMEM was applied at concentrations of 10 and 40 ng/ml on the culture dishes.

Since initial studies had shown that the optimal induction of VEGF is seen after incubation for

24 hours, this duration was chosen for the experiments. Incubation with DMEM without

PDGF, but with the same amounts of PBS/0.1% bovine serum albumin/4mM HCl, was used

as negative control.

First experiments indicated that the induction of VEGF by interleukin 1β (IL-1β) was

independent of whether the cells were cultured in DMEM/10%FCS or DMEM without FCS.

Optimal results were obtained after 6 hours of incubation. Therefore, experiments were

carried out in DMEM/10%FCS for 6 hours. Medium was changed 24 hours before the

initiation of the experiments. Recombinant IL-1β (R&D Systems) was stored in aliquots at -

80°C in stock concentration of 2 µg/ml in PBS/0.1% bovine serum albumin. After washing

with PBS, 1.5 ml DMEM/10%FCS containing recombinant IL-1β at concentrations of 1

pg/ml, 10 pg/ml and 100 pg/ml were applied on the culture dishes. Controls consisted of

DMEM/10%FCS without IL-1β, but with the same amounts of PBS/0.1% bovine serum

albumin used for the stimulation. Co-stimulation experiments were performed with 40 ng/ml

PDGF and 100 pg/ml IL-1β in DMEM without FCS for 24 hours under the same conditions as

described above.



9

Tables

Numbers Gender

(F/M)

Age

(median/range)

Disease subset

(diffuse/limited)

Disease duration

(median/range)

Skin

biopsies N = 15 12/3

44 years

(22-76) 9/6

2.6 years

(0.3 – 19)

Cultured

fibroblasts N = 5 5/0

50 years

(49-63) 3/2

3 years

(1 – 43)

PO2

measurements N = 13 12/1

62 years

(51-70) 3/10

6 years

(2-26)

Online Table 1: Characteristics of SSc patients. F = female, M = male. The disease subset

was determined according to the criteria proposed by LeRoy et al 9. Disease duration was

measured from the onset of the first non-Raynaud symptoms attributable to SSc.



10

Expression of HIF-1α

No staining (+) + ++ +++

Epidermis 0/6 0/6 1/6 3/6 2/6 Normal

Dermis 5/6 - - - -

Epidermis 1/8 4/8 2/8 1/8 0/8 SSc

Dermis 8/8 - - - -

Online Table 2: Semiquantitative analyses of the HIF-1α expression in skin specimens from

normal controls and patients with SSc. Numbers of patients/controls for each histological

score are shown. In all skin specimens from healthy controls, an expression of HIF-1α could

be detected in the epidermis. In the majority of SSc patients, the expression of HIF-1α was

weak or absent in the epidermis and no expression could be detected in the dermis,

contrasting with the abundant expression of VEGF in the same patients (see Figures 1 and 4).

(+) = weak staining of a few keratinocytes, + = <25% of keratinocytes stained positive, ++ =

25-50% positive, +++ > 50% positive



11

Expression of flt-1/VEGFR1


Normal 5/6 0/6 1/6 0/6 0/6

SSc 0/15 2/15 7/15 5/15 1/15

Expression of flk-1/VEGFR2

Normal 6/6 0/6 0/6 0/6 0/6

SSc 1/14 1/14 3/14 4/14 5/14

Online Table 3: Semi-quantitative analyses of the expression of flt-1/VEGFR1 and flk-

1/VEGFR2 in involved (fibrotic) skin biopsies from SSc patients and healthy controls.

Numbers of patients/controls for each histological score are shown. Similar to VEGF itself,

the expression of flt-1/VEGFR1 as well as flk-1/VEGFR2 showed a strong upregulation in

SSc patients compared to healthy controls (see also Figure 6 of the print-version and Figures 3

and 4 of the online supplement). (+) = weak staining of a few endothelial cells, + = <25% of

endothelial cells stained positive, ++ = 25-50% positive, +++ > 50% positive



12

Expression of flt-1/VEGFR1


SSc

involved 0/7 0/7 2/7 4/7 1/7

SSc

non-involved 0/7 2/7 4/7 0/7 1/7

Expression of flk-1/VEGFR2

SSc

involved 0/6 1/6 0/6 1/6 4/6

SSc

non-involved 0/6 1/6 3/6 2/6 0/6

Online Table 4: Semi-quantitative analyses of the expression of flt-1/VEGFR1 and flk-

1/VEGFR2 in involved skin and non-involved skin biopsies. Immunohistochemistry was

performed on fresh-frozen samples. Numbers of patients/controls for each histological score

are shown. Expression of both VEGF-Receptors was lower in non-involved (non-fibrotic)

skin biopsies than in involved (fibrotic) skin biopsies of the same patients. (+) = weak

staining of a few endothelial cells, + = <25% of endothelial cells stained positive, ++ = 25-

50% positive, +++ > 50% positive



13

Figure legends

Figure 1

Double labeling with anti-CD68 antibodies by immunohistochemistry (brown signal, AEC

development) after in situ hybridization for VEGF (dark blue signal, NBT/BCIP

development) on a skin section of a SSc patient. Coexpression with CD68 indicates that

monocytes/macrophages express VEGF mRNA in the SSc skin. Skin biopsy of clinically non-

affected skin, fresh-frozen biopsy, magnification x 630.

Figure 2

Serial sections of a skin biopsy from a patient with SSc. A, In situ hybridization showing

expression of VEGF mRNA in smaller vessels (dark blue-signal, NBT/BCIP development).

B, Coexpression with the endothelial marker von Willebrand-Factor (brown signal, Nova Red

development) in the serial section of the same biopsy indicating that VEGF mRNA is

expressed by endothelial cells. Skin biopsy of affected skin, paraffin-embedded biopsy,

magnification x 630.

Figure 3

Serial sections of a skin biopsy from another patient with SSc. A, Immunohistochemistry with

anti-flk-1/VEGFR2 antibodies showing expression of flk-1/VEGFR2 in numerous vessels

(arrows, dark blue-signal, NBT/BCIP development). B, Serial section of the same biopsy.

Immunohistochemistry with antibodies against the endothelial marker von Willebrand-Factor

(arrows, brown signal, Nova Red development). Coexpression of flk-1/VEGFR2 and von

Willebrand-Factor indicates that VEGF mRNA is expressed by endothelial cells. Skin biopsy

of affected skin, paraffin-embedded biopsy, magnification x 100.



14

Figure 4

Serial sections of a skin biopsy from an additional patient with SSc. A,

Immunohistochemistry with anti-flt-1/VEGFR1 antibodies showing expression of flt-

1/VEGFR1 in vessels (arrows, brown signal, AEC development). B, Serial section of the

same biopsy. Immunohistochemistry with antibodies against the endothelial marker von

Willebrand-Factor (arrows, dark brown signal, Nova Red development). Coexpression of flt-

1/VEGFR1 and von Willebrand-Factor indicates that VEGF mRNA is expressed by

endothelial cells. Skin biopsy of affected skin, fresh-frozen biopsy, magnification x 200.



References

1. Preliminary criteria for the classification of systemic sclerosis (scleroderma).

Subcommittee for scleroderma criteria of the American Rheumatism Association

Diagnostic and Therapeutic Criteria Committee. Arthritis Rheum. 1980; 23:581-

590.

2. Kowal-Bielecka O, Distler O, Neidhart M, Kunzler P, Rethage J, Nawrath M,

Carossino A, Pap T, Muller-Ladner U, Michel BA, Sierakowski S, Matucci-Cerinic

M, Gay RE, Gay S. Evidence of 5-lipoxygenase overexpression in the skin of

patients with systemic sclerosis: a newly identified pathway to skin inflammation in

systemic sclerosis. Arthritis Rheum.. 2001; 44:1865-1875.

3. Scheid A, Wenger RH, Christina H, Camenisch I, Ferenc A, Stauffer UG,

Gassmann M, Meuli M. Hypoxia-regulated gene expression in fetal wound

regeneration and adult wound repair. Pediat Surg Int. 2000; 16:232-236.

4. Nozue M, Lee I, Yuan F, Teicher BA, Brizel DM, Dewhirst MW, Milross CG,

Milas L, Song CW, Thomas CD, Guichard M, Evans SM, Koch CJ, Lord EM, Jain

RK, Suit HD. Interlaboratory variation in oxygen tension measurement by

Eppendorf "Histograph" and comparison with hypoxic marker. J Surg Oncol. 1997;

66:30-38.

5. Scheid A, Wenger RH, Schaffer L, Camenisch I, Distler O, Ferenc A, Cristina H,

Ryan HE, Johnson RS, Wagner KF, Stauffer UG, Bauer C, Gassmann M, Meuli M.



Physiologically low oxygen concentrations in fetal skin regulate hypoxia-inducible

factor 1 and transforming growth factor-beta3. FASEB J. 2002; 16:411-413.

6. Stroka DM, Burkhardt T, Desbaillets I, Wenger RH, Neil DA, Bauer C, Gassmann

M, Candinas D. HIF-1 is expressed in normoxic tissue and displays an organ-

specific regulation under systemic hypoxia. FASEB J. 2001; 15:2445-2453.

7. Camenisch G, Tini M, Chilov D, Kvietikova I, Srinivas V, Caro J, Spielmann P,

Wenger RH, Gassmann M. General applicability of chicken egg yolk antibodies:

the performance of IgY immunoglobulins raised against the hypoxia-inducible

factor 1alpha. FASEB J. 1999; 13:81-88.

8. Krieg M, Haas R, Brauch H, Acker T, Flamme I, Plate KH. Up-regulation of

hypoxia-inducible factors HIF-1alpha and HIF-2alpha under normoxic conditions in

renal carcinoma cells by von Hippel-Lindau tumor suppressor gene loss of function.

Oncogene. 2000; 19:5435-5443.

9. LeRoy EC, Black C, Fleischmajer R, Jablonska S, Krieg T, Medsger TA, Rowell N,

Wollheim F. Scleroderma (systemic sclerosis): classification, subsets and

pathogenesis. J Rheumatol. 1988; 15:202-205.



Fig. 1 online supplement

by guest on February 23, 2013http://circres.ahajournals.org/

Dow

nloaded from


A B



Dow

nloaded from


uncontrolled expression of vascular endothelial growth ... fileuncontrolled expression of vascular...

Documents