ORIGINAL ARTICLE

Angiogenesis in alkaptonuria

Lia Millucci1 & Giulia Bernardini1 & Barbara Marzocchi1 & Daniela Braconi1 &

Michela Geminiani1 & Silvia Gambassi1 & Marcella Laschi1 & Bruno Frediani2 &

Federico Galvagni1 & Maurizio Orlandini1 & Annalisa Santucci1

Received: 14 June 2016 /Revised: 28 July 2016 /Accepted: 17 August 2016# SSIEM 2016

Abstract Alkaptonuria (AKU) is a rare genetic disease thataffects the entire joint. Current standard of AKU treatment ispalliative and little is known about its physiopathology.Neovascularization is involved in the pathogenesis of system-ic inflammatory rheumatic diseases, a family of related disor-ders that includes AKU. Here, we investigated the presence ofneoangiogenesis in AKU synovium and healthy controls.Synovium from AKU patients, who had undergone total jointreplacement or arthroscopy, or from healthy patients withoutany history of rheumatic diseases, who underwent surgicaloperation following sport trauma was subjected to hematoxy-lin and eosin staining. Histologic grades were assigned forclinical disease activity and synovitis based on cellular contentof the synovium. By immunofluorescence microscopy, usingdifferent endothelial cell markers, we observed large vascular-ization in AKU but not in healthy synovium. Moreover,Western blotting and quantification analyses confirmed strongexpression of endothelial cell markers in AKU synovial tis-sues. Importantly, AKU synovium vascular endotheliumexpressed high levels of β-dystroglycan, a protein previouslyinvolved in the regulation of angiogenesis in osteoarthritic

synovium. This is the first report providing experimental ev-idences that new blood vessels are formed in AKU synovialtissues, opening new perspectives for AKU therapy.

AbbreviationsAKU AlkaptonuriaEC Endothelial cellvWill Von Willebrand factorVE-cad VE-cadherinβ-DG β-dystroglycanMVD Microvessel densityMVA Microvessel area

Introduction

AKU is an ultra-rare disease with no licensed therapy, eventhough a clinical trial using nitisinone is currently underway(Ranganath et al 2016). Although recent studies with an AKUmouse model contributed to clarify AKU joint pathophysiol-ogy (Taylor et al 2012), the pathological mechanisms leadingfrom homogentisic acid accumulation to tissue damage inhumans is still to be deeply investigated. Features of AKUinclude cartilage loss, osteophyte formation and synovial in-flammation. As it has been reported that angiogenesis is in-volved in a plethora of inflammatory rheumatic diseases(Maruotti et al 2013), it might contribute to AKU traits. Thesynovium plays a key role in maintaining healthy articularcartilage, which in turn modulates the function of the adjacentsynovium. Synovitis contributes not only to symptoms inAKU, but through the generation of pro-inflammatory cyto-kines (Millucci et al 2012; Spreafico et al 2013), further im-pairs chondrocyte function (Braconi et al 2012; Millucci et al2015a, b) and homeostasis of the articular cartilage (Tinti et al

Communicated by: Bruce A. Barshop

Lia Millucci and Maurizio Orlandini contributed equally to this work.

Electronic supplementary material The online version of this article(doi:10.1007/s10545-016-9976-3) contains supplementary material,which is available to authorized users.

* Annalisa Santucciannalisa.santucci@unisi.it

1 Department of Biotechnology, Chemistry and Pharmacy, Universityof Siena, Via A. Moro, 2, 53100 Siena, Italy

2 Department ofMedicine, Surgery andNeurosciences, Rheumatologysection, University of Siena, Policlinico Le Scotte, 53100 Siena, Italy

J Inherit Metab DisDOI 10.1007/s10545-016-9976-3

2010; Millucci et al 2015a, b). Moreover, macrophage infil-tration in AKU synovium has been reported (Spreafico et al2013; Millucci et al 2015a, b).

Angiogenesis is a key component of chronic inflammationand inhibition of angiogenesis can hinder synovitis in vivo(Roccaro et al 2005). Although, synovial angiogenesis hasbeen linked to synovitis by histological analyses, the potentialimportance of synovial neovascularization in AKU remainslargely unknown.

With the aim to assess the presence of neovascularization inAKU tissues, we examined the presence of different angio-genic markers in the synovium of five AKU patients aftergrading the severity of arthrosis. We observed that vonWillebrand factor (vWill), a glycoprotein produced in the en-dothelium and essential for normal hemostasis, VE-cadherin(VE-cad), an adhesion protein specific for endothelial cells(ECs), CD93, a surface protein recently involved in angiogen-esis (Orlandini et al 2014; Galvagni et al 2016), and β-dystroglycan (β-DG), a transmembrane laminin-binding pro-tein highly expressed in ECs of malignant tumors and in-flamed tissues (Hosokawa et al 2002; Wimsey et al 2006),were highly expressed in blood vessels within AKU synovialtissues. Altogether, our results suggest that synovitis is asso-ciated with increased rates of angiogenesis in AKU patients,opening new perspectives for AKU therapy.

Materials and methods

Patients

The whole study was conducted following the approval ofSiena University Hospital Ethics committee (numberGGP10058, date 07/21/2010) and has been performed in ac-cordance with the ethical standards laid down in the 1975Declaration of Helsinki and its later amendments. Five pa-tients with AKU were recruited from the section ofRheumatology, Siena University Hospital betweenMay 2013 and January 2015. The study group consisted ofthree men and two females, with an average age of 62. Nopatient had received anti-inflammatory therapy. Clinical dis-ease activity scores were calculated for patients undergoingtotal knee replacement. As a standardized and reliable clinicaland histopathological grading of AKU severity is lacking, inorder to classify AKU severity we assessed the arthritis degreeusing radiographic scores for joint space narrowing andosteophytes of the sampled knee (Online resource Table 1),as described previously (Nagaosa et al 2000). Knee radio-graphs, obtained prior to surgery, were available for the fiveAKU patients. Control patients had suffered fracture of kneerequiring surgical operation following sport trauma, and didnot showed pre-existing history of osteoarthritis symptomsand osteoarthrosis.

Bioptic samples

Serum levels of Serum Amyloid A, advanced oxidation pro-tein products, serum and urinary levels of homogentisic acidof the AKU patients were determined (Online resourceTable 2). Synovial tissues from AKU patients and healthycontrols were collected by arthroscopic biopsy. Upon openingthe joint capsule, the innermost layer of the synovium linedcapsule was harvested and processed in two different ways:(1) samples were frozen in liquid nitrogen for subsequent pro-tein extraction and Western blot analyses; (2) samples werefrozen in isopentane chilled at −196 °C for subsequent histo-logical analyses.

Histological analyses

Synovium sections (5 μm thickness) were stained withHaematoxylin and Eosin. Synovitis was assessed using a semi-quantitative score according to (Krenn et al 2002) (Onlineresource Table 1). AKU sections were morphometrically ana-lyzed for microvessel density (MVD) and microvessel area(MVA) as previously described (Weidner 1995). Briefly,microvessels were microscopically identified by the presenceof red blood cells. First, slides were examined at low magnifi-cation for identification of highly vascularized areas, then indi-vidual microvessels were counted in an adequate area (e.g.,533,448 μm2 per field using 20× objective lens and 10× ocularlens). Morphometric analyses were performed using ImageAnalysis, AxioVision BIO (Zeiss International) analysis soft-ware. Two cases of normal synovium were used as control.

Immunofluorescence analysis

Cryostat synovium sections (3 μm thickness) were fixed in4 % paraformaldehyde at room temperature. Sections wererinsed in PBS for 5 min, and then blocked using 3 % bovineserum albumin. The following primary antibodies were used:mouse anti-CD93 (clone 4E1, 0.6 mg/mL) (Orlandini et al2014); rabbit anti-vWill (A0082 Dako, Glostrup, Denmark);goat anti-VE-cad (sc-6458) and rabbit anti-β-DG (sc-28535Santa Cruz Biotechnologies, Santa Cruz, CA, USA). Sectionswere rinsed three times in PBS for 5 min and incubated withAlexa Fluor-488-conjugated secondary antibodies (ThermoFisher Scientific, Waltham, MA, USA). Slides were mountedinMowiol 4–88 and fluorescent images were captured using aLeica TCS SP2 laser-scanning confocal microscope.

Western blotting

Frozen synovial fragments were weighed and immediatelyplaced in a chilled Dounce homogenizer. Extracting buffer(9 M urea, 4 % CHAPS, 65 mM 1,4-dithioerythritol, 35 mMTris base) was added at a ratio of 50 μL per 10 mg of tissue.

J Inherit Metab Dis

First, the mixture was homogenized alternating 3 min of grind-ing with 1 min of rest in an ice bath ten times. Then, a seconddisrupting step was achieved by vortexing for 10 min, alternat-ing 30 s of vortexing with 30 s of rest in an ice bath. Finally,samples were sonicated for 30 min and centrifuged for 20 minat 17,000×g at 4 °C. The supernatant was collected and itsprotein concentration determined by Bradford’s method;40 μg of protein lysate were resolved by 8 % SDS-PAGE.Immunoblotting analyses were performed as previously de-scribed (Orlandini et al 2014). The following primary antibod-ies were used: anti-CD31 (805-003A-C100 Enzo Life Sciences,Farmingdale, NY, USA); goat anti-VE-cad and rabbit anti-β-DG; anti-β-actin (A5441 Sigma-Aldrich, St Louis, MO, USA).

Statistical analysis

Statistical evaluation was performed using Student’s t-test andvalues of p < 0.05 were considered statistically significant.

Results

Macroscopic examination of the synovial membrane fromknees of AKU patients revealed the presence of hypertrophictissue and numerous pigmented villi, some of which hyper-emic (Fig. 1a). To investigate whether angiogenesis is

occurring in AKU inflamed synovial tissues, we analyzedthe synovium from AKU patients and healthy controls byhistology. In classical inflammatory arthritides the synovialmembrane is characterized by wide heterogeneity. We ob-served that the synovium of all AKU patients also displayedthe same spectrum of changes and showed evidence of severe(grade 3) synovial inflammation with histologic features typ-ical of acute synovitis (Online resource Table 1). Hyperplasiaof the lining layer, increased vascularity, abundance of carti-lage shards in the upper sublining layer, and inflammatory cellinfiltration were found typical of AKU synovium (Fig. 1b–d).Pannus formed and the presence of blood vessels, neutrophils,lymphocytes, and hyperplastic synoviocytes emerged(Fig. 1d), suggesting that angiogenesis is arising in AKUsynovium.

To address whether neovascularization is a trait of AKUsynovial tissues, we performed immunofluorescence analysesusing different EC markers, including β-DG, which is highlyexpressed during formation of new blood vessels in osteoar-thritic synovium (Wimsey et al 2006). Imaging of synovialtissues showed that, in addition to vWill and VE-cad, β-DGand CD93 were highly expressed in blood vessels of AKUsynovia in comparison to healthy controls, suggesting that inthese inflamed tissues new blood vessels are forming (Fig. 2).Areas of high vessel density, spaced out by areas with lowervessel content, were detected in all samples of AKU synovium

Fig. 1 AKU synovial tissues exhibit inflammation and vascularization.Cryostat sections were stained with haematoxylin and eosin andrepresentative images are shown. a Picture showing gross pathologicAKU synovium. Ochronotic pigment is visible in black areas. Somehyperemic pigmented villi are indicated by arrows. b Section of aseverely degraded AKU synovial membrane. Hyperplasia of the lininglayer and regions of predominantly lymphocytic inflammation are

indicated by arrows. An ochronotic cartilage shard is indicated by anasterisk. Magnification, 20×. c Synovial capsular tissue with a highgrade of vascularization is shown. Arrows indicate blood vessels.Magnification, 20×. d Section showing large shards (asterisks)embedded in synovial tissue, surrounded by macrophages and giantcells (arrowheads). Blood vessels were mainly concentrated in theproximity of ochronotic shards (arrows). Magnification, 40×

J Inherit Metab Dis

and increased with increasing inflammation grade. To quanti-tatively analyze synovial angiogenesis, number of bloodvessels/field and percentages of vascularized area were count-ed both in histological and immunofluorescence stained sec-tions. MVD and MVAwere significantly increased in the sy-novial pannus of AKU patients with respect to healthy con-trols (Fig. 3a–c and Online resource Fig. 1). Finally, formationof new blood vessels was confirmed byWestern blotting anal-yses, showing strong expression of EC markers in AKU sy-novia while in healthy synovia their expression was faint(Fig. 3d).

Discussion

Subclinical inflammation is common in AKU and circu-lating markers of inflammation are elevated in AKU

populations (Tinti et al 2010; Laschi et al 2012; Mistryet al 2015). In the present work we showed for the firstt ime tha t AKU was accompan ied by synov ia lneoangiogenesis, suggesting that together with inflam-mation it is directly implicated in the initiation and pro-gression of AKU. Indeed, we demonstrated the co-presence of inflammatory and neoangiogenic evidencein AKU synovial membranes compared to healthy con-trols. Since β-DG and CD93 expression levels werefound to be high in ECs of malignant tumors, hyperplas-tic and inflamed tissues, where endothelial cells wereassembling new blood vessels (Wimsey et al 2006;Langenkamp et al 2015; Galvagni et al 2016), our resultsshowing high expression of β-DG and CD93 proteins inECs of AKU inflamed synovia suggest new blood vesselformation in this tissue. Synovial neoangiogenesis andinflammation were observed in all AKU patients

Fig. 2 Newly-formed bloodvessels in AKU synovial tissues.Cryostat synovium sections werestained by immunofluorescenceusing anti-vWill, anti-VE-cad,anti-CD93, and anti-β-DGantibodies. Representative imagesof two patients and a healthycontrol are shown. Scale bars,150 μm

J Inherit Metab Dis

presenting a high grade of articular degeneration, al-though they could not be limited to late-stage disease(Millucci et al 2012, 2015a, b).

Blood vessel growth is a key factor in progression tochronic rheumatic diseases. During synovitis, the expandedvascular network supplies the large amount of nutrients andoxygen required by the growing synovial mass and alsoprovides an easy access of inflammatory cells and pro-inflammatory cytokines (McInnes and Schett 2007). In ad-dition, the proliferating synovium creates hypoxic regionsthat, together with the release of pro-inflammatory factors,acts as a powerful pro-angiogenic stimulus (Khurana et al2005). Angiogenesis may also indirectly promote itself byincreasing inflammatory cell infiltration, thereby increasingthe availability of angiogenic growth factors produced bythese cells. Finally, during early synovitis, angiogenesismay contribute to the transition from acute to chronic in-flammation (Walsh 1999). Therefore, in AKU synovia, thecrosstalk between angiogenesis and inflammation could

contribute to the damage of the osteoarthritic joint throughcartilage degradation and osteophyte formation (Bonnetand Walsh 2005).

Pharmaceutical agents designed to modify the progress ofinflammation in rheumatoid arthritis have largely been devel-oped. The intensive search for anti-angiogenic agents has beendriven by therapeutic potential in oncology. Although, the mo-lecular crosstalk between inflammation and angiogenesis maydiffer between AKU and cancer, we cannot exclude the appli-cation of existing anti-inflammatory and/or antiangiogenic ther-apies to diseases that were previously thought of as ‘degenera-tive’. However, the greatest therapeutic potential will comefrom finding the exact disease-specific molecular mechanismsthat drive inflammation and angiogenesis.

Acknowledgments We thank Dr. Dario Gambera and Prof. Pier PaoloMariani for assisting in sample collection and AIMAKU (AssociazioneItaliana Malati di Alkaptnoria) for its support. This work was funded byTelethon Italy (grant GGP10058).

Fig. 3 Quantification of immunofluorescence stained synovial sections.a Images shown in Fig. 2 and displaying vWill, CD93, VE-cad, and β-DG staining distribution in AKU and healthy (ctr) synovia werequantitatively analyzed as individual channels by using ImageJsoftware. Single blood vessels were selected as regions of interest andbackground signal was subtracted. Representative analyses of AKUpatient and healthy control are reported. b Plot showing the percentageof pixel color intensity distribution of the acquired images analyzed as in

(a) for five AKU patients and two controls. c MVD and percentage ofstained area were analyzed in synovial tissue sections from five AKUpatients and two controls (ctr) stained with anti-vWill antibodies.Results represent the means ± SD of three independent experiments(p < 0.01). d Cell extracts from AKU and healthy synovia wereanalyzed by Western blotting using anti-CD31, anti-VE-cad, anti-β-DG, and anti-β-actin antibodies to confirm equal loading. Arepresentative experiment is shown

J Inherit Metab Dis

Compliance with ethical standards

Conflicts of interest None.

Study in human subjects All procedures followed were in accordancewith the ethical standards of the responsible committee on human exper-imentation (institutional and national) and with the Helsinki Declarationof 1975, as revised in 2000. Informed consent was obtained from allparents of patients for being included in the study.

References

Bonnet CS,Walsh DA (2005) Osteoarthritis, angiogenesis and inflamma-tion. Rheumatology 44:7–16

Braconi D, Bernardini G, Bianchini C et al (2012) Biochemical andproteomic characterization of alkaptonuric chondrocytes. J CellPhysiol 227:3333–3343

Galvagni F, Nardi F, Maida M et al (2016) CD93 and dystroglycan co-operation in human endothelial cell adhesion and migration.Oncotarget 7:10090–10103

Hosokawa H, Ninomiya H, Kitamura Y, Fujiwara K, Masaki T (2002)Vascular endothelial cells that express dystroglycan are involved inangiogenesis. J Cell Sci 115:1487–1496

Khurana R, Simons M, Martin JF, Zachary IC (2005) Role of angiogen-esis in cardiovascular disease: a rritical appraisal. Circulation 112:1813–1824

Krenn V, Morawietz L, Häupl T, Neidel J, Petersen I, König A(2002) Grading of chronic synovitis—a histopathological grad-ing system for molecular and diagnostic Pathology. Pathol ResPract 198:317–325

Langenkamp E, Zhang L, Lugano R et al (2015) Elevated expression ofthe C-type lectin CD93 in the glioblastoma vasculature regulatescytoskeletal rearrangements that enhance vessel function and reducehost survival. Cancer Res 75:4504–4516

Laschi M, Tinti L, Braconi D et al (2012) Homogentisate 1,2 dioxygenaseis expressed in human osteoarticular cells: implications inalkaptonuria. J Cell Physiol 227:3254–3257

Maruotti N, Annese T, Cantatore FP, Ribatti D (2013) Macrophages andangiogenesis in rheumatic diseases. Vasc Cell 5:1–8

McInnes IB, Schett G (2007) Cytokines in the pathogenesis of rheuma-toid arthritis. Nat Rev Immunol 7:429–442

Millucci L, Spreafico A, Tinti L et al (2012) Alkaptonuria is a novelhuman secondary amyloidogenic disease. Biochim Biophys Acta1822:1682–1691

Millucci L, Braconi D, Bernardini G et al (2015a) Amyloidosis inalkaptonuria. J Inherit Metab Dis 38:797–805

Millucci L, Giorgetti G, Viti C et al (2015b) Chondroptosis inalkaptonuric cartilage. J Cell Physiol 230:1148–1157

Mistry JB, Jackson DJ, Bukhari M, Taylor AM (2015) A role for inter-leukins in ochronosis in a chondrocyte in vitro model ofalkaptonuria. Clin Rheumatol

Nagaosa Y, Mateus M, Hassan B, Lanyon P, Doherty M (2000)Development of a logically devised line drawing atlas for gradingof knee osteoarthritis. Ann Rheum Dis 59:587–595

OrlandiniM, Galvagni F, Bardelli M et al (2014) The characterization of anovel monoclonal antibody against CD93 unveils a newantiangiogenic target. Oncotarget 5:2750–2760

Ranganath LR, Milan AM, Hughes AT et al (2016) Suitability ofnitisinone in alkaptonuria 1 (SONIA 1): an international,multicentre, randomised, open-label, no-treatment controlled, paral-lel-group, dose–response study to investigate the effect of once dailynitisinone on 24-h urinary homogentisic acid excretion in patientswith alkaptonuria after 4 weeks of treatment. Ann Rheum Dis 75:362–367

Roccaro A, Francesca RF, Teresa CT, Di Pietro G, Vacca A, Dammacco F(2005) Antiangiogenesis for rheumatoid arthritis. Curr Drug TargetsInflamm Allergy 4:27–30

Spreafico A, Millucci L, Ghezzi L et al (2013) Antioxidants inhibit SAAformation and pro-inflammatory cytokine release in a human cellmodel of alkaptonuria. Rheumatology 52:1667–1673

Taylor AM, Preston AJ, Paulk NK et al (2012) Ochronosis in a murinemodel of alkaptonuria is synonymous to that in the human condi-tion. Osteoarthr Cartil 20:880–886

Tinti L, Spreafico A, Braconi D et al (2010) Evaluation of antioxiodantdrugs for the treatment of ochronotic alkaptonuria in an in vitrohuman cell model. J Cell Physiol 225:84–91

Walsh DA (1999) Angiogenesis and arthritis. Rheumatology 38:103–112Weidner N (1995) Current pathologic methods for measuring

intratumoral microvessel density within breast carcinoma and othersolid tumors. Breast Cancer Res Treat 36:169–180

Wimsey S, Lien CF, Sharma S et al (2006) Changes inimmunolocalisation of β-dystroglycan and specific degradativeenzymes in the osteoarthritic synovium. Osteoarthr Cartil 14:1181–1188

J Inherit Metab Dis