Small leucine-rich proteoglycans, decorin and fibromodulin, are reduced in postburn hypertrophic...

Small leucine-rich proteoglycans, decorin and fibromodulin,are reduced in postburn hypertrophic scar

Dariush Honardoust, PhD; Mathew Varkey, MSc; Keijiro Hori, MD; Jie Ding, MD, PhD; Heather A. Shankowsky,RN; Edward E. Tredget, MD, MSc

Wound Healing Research Group, Department of Surgery, Division of Plastic and Reconstructive Surgery, University of Alberta, Edmonton, Alberta,

Canada

Reprint requests:Edward E. Tredget, Department of Surgery,

Division of Plastic and Reconstructive

Surgery, University of Alberta, 2D2.28

WMC, 8440-112 Street, Edmonton AB,

Canada T6G 2B7.

Tel: 1780 407 6979;

Fax: 1780 407 7394;

Email: [email protected]

Manuscript received: May 20, 2010

Accepted in final form: January 25, 2011

DOI:10.1111/j.1524-475X.2011.00677.x

ABSTRACT

Small leucine-rich proteoglycans (SLRPs) are extracellular matrix molecules thatregulate collagen fibrillogenesis and inhibit transforming growth factor-b activ-ity; thus, they may play a critical role in wound healing and scar formation.Hypertrophic scarring is a dermal form of fibroproliferative disorders, whichoccurs in over 70% of burn patients and leads to disfigurement and limitationsin function. By understanding the cellular and molecular mechanisms that leadto scarring after injury, new clinical therapeutic approaches can by developed tominimize abnormal scar formation in hypertrophic scarring and other fibropro-liferative disorders. To study the expression and localization of SLRPs withconnective tissue cells in tissue immunohistochemistry, immunofluorescencestaining, immunoblotting, and reverse-transcription polymerase chain reactionwere used in normal skin and hypertrophic scar (HTS). In normal skin, there wasmore decorin and fibromodulin accumulation in the superficial layers than in thedeeper dermal layers. The levels of decorin and fibromodulin were significantlylower in HTS, whereas biglycan was increased when compared with normal skin.There was an increased expression of biglycan, fibromodulin, and lumican in thebasement membrane and around basal epithelial cells. In contrast, these pro-teoglycans were absent or weakly expressed in HTS. The findings suggest thatdown-regulation of SLRPs after wound healing in deep injuries to the skin playsan important role in the development of fibrosis and HTS.

After burn injury, > 70% of the patients develop hyper-trophic scars (HTS) that can result in tissue disfigurementor disruption of normal function. A better understandingof the differences in the cellular andmolecular mechanismsinvolved in normal wound healing and the development ofHTSmay provide valuable information that can be used toprevent and treat abnormal scar formation. Wound heal-ing is a process of continuous repair, in which differentcells such as platelets, fibroblasts, epithelial, endothelial,and inflammatory cells interact with extracellular matrix(ECM) molecules, growth factors, and cytokines. Aberra-tions in this process may result in chronic nonhealingwounds or deposition of excess collagen and HTS forma-tion. Small leucine-rich proteoglycans (SLRPs) decorin,biglycan, fibromodulin, and lumican are structurally andfunctionally related ECM molecules that are abundantlypresent in connective tissue and regulate important cellfunctions involved in wound healing including gene ex-pression, cell adhesion, and growth factor activity.1,2 Animportant function of SLRPs is their ability to bind type Icollagen and regulate collagen fibrillogenesis. It is knownthat SLRPs are expressed in close association with differ-ent connective tissue cells suggesting the cells producethese molecules and interact with them.3,4 Consideringthe properties of SLRPs in the regulation of cell functionand collagen fibrillogenesis, they likely play a key role inwound healing and scar formation.

The deposition and turnover of collagen ultimatelyresults in HTS formation but the changes in expression,

localization, and type of proteoglycans are importantfactors in the abnormal morphology of the ECM and thealtered physical properties of HTS5 such as excessiveaccumulation of thick collagen fiber bundles containing ahigher density of endothelial cells (EC), fibroblasts, andmyofibroblasts.6 Although fibroblasts are the major celltype associated with fibrillogenesis and wound healing,myofibroblasts may be an important source of ECM in thedevelopment of fibrosis.7 The transition from fibroblaststo myofibroblasts is influenced by transforming growthfactor-b1 (TGF-b1), a key profibrotic mediator and animportant regulator of wound healing cellular processesincluding proliferation, differentiation, migration, cellsurvival, and angiogenesis. Decorin, biglycan, and fibro-modulin can bind to and inhibit TGF-b1 activity invitro8 and in vivo.9,10 For example, fibromodulin showedan inverse relationship with TGF-b1 activity during scarformation in both fetal and adult rat wound repair.9

In addition, adenovirus-mediated overexpression of fib-romodulin improved wound healing in incisional woundsand reduced scarring.11 Recombinant decorin reducedcollagen and TGF-b1 synthesis in cultured HTS fibro-blasts.12,13 In contrast, overexpression of TGF-b1 resultedin marked lung fibrosis, which was significantly antago-nized by parallel overexpression of decorin;10 therefore,the hypothesis is that the decreased expression of SLRPs atspecific time points during wound healing in the skinfollowing a burn injury leads to the development of HTS.Changes in the distribution and lower expression of

Wound Rep Reg (2011) 19 368–378 c� 2011 by the Wound Healing Society368

Wound Repair and Regeneration

mailto:[email protected]

SLRPs in deep dermal wounds may result in an up-regulation of TGF-b1 activity and increased collagensynthesis and deposition. In this study, differences in theexpression, cell-association, and localization of SLRPs innormal skin and HTS were analyzed. These findingsmay contribute to the characterization of the cellular andmolecular mechanism of SLRPs in the development ofHTS following burn injury.

MATERIALS AND METHODS

Preparation of tissue samples

Protocols for human tissue sampling have been approvedby the University of Alberta Hospitals Health ResearchEthics Board. Patients included in the study are patientstreated at the University of Alberta Hospital, Firefighters’Burn Treatment Unit. Following informed consent,normal skin and HTS tissue samples were obtained frompatients recovering from thermal injury that developedclinical characteristics of HTS including thickened, red,raised, and itchy scars confined to the site of injury. Site-matched normal skin from the same patients was used ascontrols. A description of the patient samples used includ-ing gender, extent of injury, time postinjury, and locationof biopsies are summarized in Table 1.

Antibodies

To analyze the protein expression level, localization, andabundance of target molecules, the following antibodieswere used for immunoblotting and immunostaining: poly-clonal anti-decorin from R&D Systems (Minneapolis,MN), monoclonal anti-decorin (9XX, sc-73896), poly-clonal anti-biglycan (H-150, sc-33788), polyclonal anti-fibromodulin (H-50, sc-33772), and polyclonal anti-lumican(H-90, sc-33785) from Santa Cruz Biotechnology (SantaCruz, CA). To identify different connective tissue cells,monoclonal antibodies that recognize specific molecules

expressed by myofibroblasts and pericytes (a-smooth mus-cle actin [a-SMA]; NeoMarkers, Fremont, CA) and EC(CD31; NeoMarkers) were used. Appropriate Alexa-con-jugated anti-rabbit-594 or anti-mouse-488 (MolecularProbes Inc., Eugene, OR) was used as the secondary anti-body. For Western blot loading controls, polyclonalanti-b-actin from Sigma (St. Louis, MO) was used. Goatanti-rabbit and anti-mouse IgG horse-radish peroxidase(HRP) conjugate secondary antibodies used for immuno-blotting were purchased from BioRad (Hercules, CA).

Immunofluorescence staining and confocal microscopy

For immunofluorescence staining, paraffin-embedded tis-sue sections were deparaffinized in two washes of xylenefor 5 minutes and hydrated gradually through 100%, 90%,80%, and 70% alcohol for 5 minutes each. For antigenunmasking, tissue sections were exposed to 0.05% saponinin deionized water at room temperature for 30 minutes.Tissue sections were then washed three times with phos-phate-buffered saline (PBS), blocked with PBS containingbovine serum albumin (BSA; 10mg/mL), and Triton X-100 (0.01%) for 1 hour at room temperature, then incu-bated with the primary antibody diluted in PBS containingBSA (1mg/mL) and Triton X-100 (0.01%) overnight at4 1C. The tissue sections were then washed and incubatedwith the appropriate secondary antibody for 1 hour atroom temperature before mounting using Immuno-mountsolution (Thermo Shadon, Pittsburgh, PA). For doubleimmunofluorescence staining, the tissue sections were in-cubated with the first primary antibody as above. Sectionswere then washed and incubated with the second primaryantibody for 1 hour at room temperature followed by1-hour incubation with the appropriate Alexa-conjugatedanti-rabbit-594 or anti-mouse-488. Images were capturedusing Carl Ziess Laser Confocal Microscope equippedwith LSM5 software (Carl Ziess MicroImaging Inc.,Thornwood, NY). Control immunostaining was per-formed using an appropriate nonimmune serum resultingin negative staining results (data not shown).

Isolation of fibroblasts from normal skin and HTS

Briefly, normal skin and HTS tissue samples were mincedinto small pieces (2–3mm3) and digested in Dulbecco’smodified Eagle’s medium (DMEM; Gibco, Invitrogen,Burlington, ON, Canada) supplemented with 10% FBScontaining Clostridium histolyticum collagenase (CHCextract) (4mL/g tissue), gently shaking (0.316–0.382 � g)for 2 hours at 37 1C. Digested tissue was filtered usingan open filter chamber (NPBI, Emmer-Compascuum,the Netherlands). Cells were collected by centrifugation(182 � g for 5 minutes) and isolated cells were cultured in75 cm2 culture flasks (Corning Costar Corp., Cambridge,MA) in DMEM supplemented with 10% FBS. Strains ofnormal dermal and HTS fibroblasts at passage 2–5 wereused in this study.

Western blotting

To analyze decorin, biglycan, fibromodulin, and lumicanexpression in normal skin and HTS, individual tissue sam-ples were minced (2–3mm3), placed into vials containing

Table 1. Patient demographic information

Patient Gender Agen TBSA (%)w Type of scar Location Timez

1 M 38 35 Burn HTS Hand 74


3 F 26 No burn Incision HTS Shoulder 257

4 M 35 36 Burn HTS Arm 20

5 F 60 No burn Acne HTS Chest 467

6 M 38 7 Burn HTS Arm 45

7 M 37 2 Burn HTS Eyelid 244




11 M 9 25 Burn HTS Chest 74

nAge at injury.wTotal body surface area (TBSA%) refers to the extent of burn

injury.zWeeks postwounding when samples were collected.

Wound Rep Reg (2011) 19 368–378 c� 2011 by the Wound Healing Society 369

SLRP, decorin, fibromodulin, are reduced in HTSHonardoust et al.

ball bearings, deep frozen in liquid nitrogen and shaken topowder at 854 � g using a Micro-dismembrator (B. BraunBiotech Inc., Allentown, PA). One milliliter of lysis buffer(PBS, Triton X-100, EDTA, protease inhibitor cocktail)was added to each vial. The dissolved tissue was trans-ferred to 1.5mL Eppendorff tubes and left on ice for 30minutes and vortexed intermittently, then centrifugedat 24,770 � g for 15 minutes at 4 1C. Supernatants werecollected and the protein concentration was determinedusing a Bradford protein assay. To determine proteinexpression levels of SLRPs in cultured fibroblasts, cellswere grown to 95% confluency in DMEM/10% FBS orserum-free DMEM in 25mL culture flasks. After lysingthe cells by adding 1mL of lysis buffer to each flask, pro-tein concentrations were measured using a Bradford pro-tein assay. Equal amounts of solubilized protein fromtissue or cultured fibroblasts were electrophoresed in 10%sodium dodecyl sulfate (SDS)-polyacrylamide gels in0.1M tris borate/SDS buffer. The proteins were trans-ferred to PVDF membranes and incubated with primaryantibodies overnight at 4 1C, followed by 1-hour incubationwith appropriate HRP-conjugated antibody (BioRad). An-tibody binding was probed using a chemiluminescence de-tection system (PerkinElmer Life Sciences, Boston, MA)and visualized by exposing membranes to Kodak x-ray film(BioMax, Rochester, New York, NY). The blots werescanned and the band intensity was quantified using ImageJ software (http://rsb.info.nih.gov/ij/). Statistical analysiswas performed by paired Student’s t test for differences inthe means of sample pairs. Values of p < 0.05 were consid-ered to be statistically significant.

Real-time reverse transcription-polymerase chain

reaction (RT-PCR)

Total RNA was extracted from the normal skin and HTStissue using RNeasy spin columns (Qiagen, Mississauga,ON, Canada) according to the manufacturer’s recommen-dations. Real-time RT-PCR was conducted using PowerSYBRs Green PCRMaster Mix (ABI, Foster City, CA) ina 25mL tube with a total reaction volume of 25mL contain-ing 5mL of a 1 : 2 dilution of first-strand reaction product,0.4mM gene-specific upstream and downstream primers.Amplification and analysis of cDNA fragments were carriedout using a StepOne Plus real-time PCR system (AppliedBiosystems, Foster City, CA). Cycling conditions with ini-tial denaturation at 95 1C for 3 minutes, followed by 40cycles consisting of a 15-second denaturation intervalat 95 1C and a 30-second interval for annealing andprimer extension at 60 1C. Amplification of the housekeep-ing gene hypoxanthine–guanine–phosphoribosyl transferase(HPRT) mRNA, which served as a normalization standard,was carried out with HPRT primers GACCAGT CAACAGGGGACA (forward) and ACACTTCGTGGGGTCCTTTT (reverse). The gene-specific primers for decorin wereTGTCATAGAACTGGGCACCAAT (forward) andGGAAAGCCCCATTTTCAATTC (reverse), biglycan TCTCCAGCACCTCTACGTAAGGA (forward) and ACCCCTGCCCTCTGCTGTA (reverse), fibromodulin TTTTATCATCGTTCTGCCTTCATG (forward) and TGTTTGCGGGACCTTAGGAA (reverse), and lumican GCAATCGCATCTCAGAAACCA (forward) and TCGTTAGCAAC

ACGTAGACATTCA (reverse). Levels of mRNA weremeasured as CT threshold levels and normalized with theindividual HPRT control CT values.

RESULTS

Specimens were obtained from patients recovering fromburn injury that showed clinical signs of HTS includingthickened, red, raised, and itchy scars confined to the siteof injury. Normal skin samples from the same patientswere used as controls. At the cellular level, histologicalcharacteristics of HTS compared with normal skin areshown in Figure 1. Compared with the epithelium ofnormal skin (Figure 1A, double-headed arrow), HTSscar showed a thicker epithelium and higher cell density(Figure 1B, double-headed arrow). In normal skin, thincollagen fibers were organized in a basket-weave orienta-tion throughout the connective tissue (Figure 1C). Com-pared with deep areas of normal skin (Figure 1A and G,arrows), the papillary and superficial connective tissueshowed decreased accumulation of collagen and thinnerfibers (Figure 1A and G, arrowheads). In contrast, inthe HTS tissue, loosely packed collagen fiber bundlesappeared to be thicker with a parallel noodle-shapedstructure oriented in one direction, indicating contraction(Figure 1D). Normal skin contained few blood vessels perunit field (Figure 1E, arrows); however, in HTS tissuethere was a remarkable increase in the number of bloodvessels (Figure 1F, arrows). HTS also showed strikinghypercellularity (Figure 1H) compared with normal skin(Figure 1G).

Immunofluorescence staining of each of the SLRPsshowed that immunoreactivity of decorin in HTS (Figure2B) was considerably lower than in normal skin (Figure2A). Within the normal skin, there were lower concentra-tions of decorin in the deeper areas of connective tissue(Figure 2A, Deep) compared with the papillary and super-ficial areas (Figure 2A, SF). Lower expression of decorinin the deep dermis coincided with deposition of thicker anddenser collagen fibers in this area (Figure 1A and G,arrows). These data suggest a correlation between lowerexpression of decorin and increased accumulation ofthicker collagen fibers during wound healing. Clinically,injuries that affect the deeper layers of the dermis showmore pronounced scar formation than is seen in superficialinjuries. There is increased staining intensity of biglycan inHTS (Figure 2D) than normal skin (Figure 2C); however,in normal skin, biglycan accumulation was observed in thebasement membrane (Figure 2C, arrows) but absent inthese same regions in HTS (Figure 2D). Fibromodulin ex-pression was remarkably higher in normal skin (Figure 2E)compared with HTS tissue (Figure 2F). Likewise, accumu-lation of fibromodulin in basal epithelial cells and thebasement membrane appeared to be increased in normalskin (Figure 2E, arrows) compared with HTS (Figure 2F,arrows). The immunoreactivity of lumican was slightlylower in HTS (Figure 2H) as compared with normal skin(Figure 2G), especially in the basement membrane andpapillary and superficial areas of the connective tissue(Figure 2G and H, arrows). Consistent with these findingsusing immunostaining, Western blotting analysis ofSLRPs revealed that protein expression levels of decorinand fibromodulin were significantly down-regulated in


SLRP, decorin, fibromodulin, are reduced in HTS Honardoust et al.

http://rsb.info.nih.gov/ij/

HTS tissue compared with normal skin (Figure 3A and C).No significant changes were observed for protein expres-sion of biglycan or lumican in normal skin and HTS (Fig-ure 3B and D). In quantitative RT-PCR, the HTS hadconsiderably lower decorin and fibromodulin mRNA lev-els compared with normal skin whereas, biglycan, andlumican were significantly higher in HTS when comparedwith normal skin (Figure 3E).

Previous data have indicated that SLRPs colocalized withdifferent cells in wounds and normal skin tissues in a modelof nonscarring human gingival wound healing, where eachof the SLRPs showed distinct association with capillary EC,a-SMA–expressing cells around the blood vessels (pericytes)

or myofibroblasts.3,4 In order to investigate the associationbetween SLRPs and HTS cells, localization and immuno-reactivity of decorin, biglycan, fibromodulin, and lumicanwere assessed using double immunofluorescence staining.Decorin colocalized with some a-SMA–expressing cells(Figure 4A–C, arrowheads) and was strongly expressed bya-SMA–expressing blood vessels (pericytes) in normal skin(Figure 4A–C, arrows) while it was absent or weaklyexpressed in pericytes of HTS tissue (Figure 4D–F, arrows).Although HTS showed a dramatic increase in the numberof myofibroblasts (Figure 4G), relative to normal skin, it isunclear whether or not a-SMA–expressing cells in normalskin (Figure 4A, arrowheads) are myofibroblasts. Further

Figure 1. Histological evaluation of

hypertrophic scar tissue and normal

skin by H&E staining. A comparison

of epithelium thickness (A and B),

collagen orientation and organization

(C and D), number of blood vessels

per unit field (E and F) and cell

density (G and H) between normal

skin (A, C, E, and G) and hypertro-

phic scar tissue (B, D, F, and H).

Double-headed arrows in (A) and (B)

represent average thickness of epi-

thelium. Arrowheads and arrows in

(A) and (G) point to collagen in the

superficial and deep dermis, respec-

tively. Arrows in (E) and (F) point to

blood vessels. The arrowhead in (F)

points to red blood cells. Scale bars:

50 mm. H&E, hematoxylin & eosin.



investigation to characterize these cells is needed. Decorincolocalized with some myofibroblasts in HTS tissue (Figure4G–I). In normal skin, CD311 EC and blood vesselsstrongly expressed decorin (Figure 4J–L, arrowheads andarrows, respectively), whereas CD311 blood vessels in HTS

showed weak or negative immunoreactivity to decorin (Fig-ure 4M–O, arrows). Interestingly, decorin colocalized withmigrating CD311 EC in HTS (Figure 4M–O, arrowheads).As well, in normal skin, fibromodulin was strongly ex-pressed in a-SMA–expressing blood vessels (Figure 5A–C

Figure 2. Evaluation of immunofluorescence staining intensity

and localization of SLRPs in HTS tissue and normal skin: Decorin

in normal skin and HTS, respectively (A, B). Biglycan in normal skin

and HTS, respectively (C, D). Fibromodulin in normal skin and HTS,

respectively (E, F). Lumican in normal skin and HTS, respectively

(G, H). Lines in A separate superficial and deep connective tissue.

Arrows point to areas immediately below the epithelium and base-

ment membrane. Ep, epithelium; SF, superficial; HTS, hypertrophic

scar; SLRPs, small leucine-rich proteoglycans. Scale bar: 50mm

Figure 3. Western blotting and quantitative RT-PCR analysis of

SLRPs in normal unwounded skin and HTS. Protein expression

level of decorin (A), biglycan (B), fibromodulin (C), and lumican (D)

and mRNA expression level of SLRPs (E) in HTS and site-matched

normal unwounded skin from the same patient. N, normal

unwounded skin; HTS, hypertrophic scar; DCN, decorin; BGN,

biglycan; FMN, fibromodulin; LMN, lumican; RT-PCR, reverse

transcription-polymerase chain reaction; SLRPs, small leucine-rich

proteoglycans; np < 0.05 (one-way ANOVA and Dunnett’s

multiple comparison test); b-actin represents loading control.



arrows), but was absent or weakly expressed in pericytesof HTS capillaries (Figure 5D–F arrows). In HTS tissue,a-SMA–expressing EC that appeared to be migrating to thesite of neovascularization (Figure 5D–F, arrowheads),

myofibroblasts (Figure 5G–I, arrows) and other a-SMA–expressing cells (Figure 5g–i) showed strong staining inten-sity for fibromodulin. CD311 blood vessels in normal skinshowed positive staining for fibromodulin (Figure 5J–L,

Figure 4. Localization of decorin in

normal skin and HTS cells. Repre-

sentative images from normal skin

and HTS. a-SMA–expressing blood

vessel in normal skin and HTS,

respectively (A, D). Decorin staining

in normal skin and HTS, respectively

(B, E), and corresponding merged

images (C, F) (yellow indicates colo-

calization). Arrows indicate blood ves-

sels; arrowheads indicate a-SMA–

expressing cells. Cell-associated stain-

ing for a-SMA and decorin, respec-

tively, in HTS (G, H) and corresponding

merged images (I); yellow indicates co-

localization. Staining for CD311 endo-

thelial cells and blood vessels in normal

skin and HTS, respectively (J, M). Stain-

ing for decorin in normal skin and HTS,

respectively (K, N), and corresponding

merged images (L, O) (arrows point to

blood vessels; arrowheads indicate

endothelial cells). Yellow indicates co-

localization. HTS, hypertrophic scar;

a-SMA; a-smooth muscle actin. Scale

bars: 50mm.



arrows), whereas they did not show immunoreactivityfor fibromodulin inHTS (Figure 5M–O, arrowheads). How-ever, CD311 EC migrating to the site of neovascularizationweakly expressed fibromodulin (Figure 5M–O, arrows).

Biglycan and lumican were colocalized with CD311 anda-SMA–expressing blood vessels and cells with no signifi-cant difference seen between normal skin and HTS (datanot shown).

Figure 5. Localization of fibromodulin

in normal skin and HTS cells. Repre-

sentative images from normal skin

and HTS. Staining for a-SMA and

fibromodulin, respectively, in normal

skin (A, B). Corresponding merged im-

age (C) (arrows point to blood ves-

sels). Staining for a-SMA and

fibromodulin, respectively, in HTS (D,

E) (arrows indicate blood vessels;

arrowheads indicate a-SMA–express-

ing cells). Corresponding merged

image (F); yellow indicates colocal-

ization. Staining for a-SMA and fib-

romodulin, respectively, in HTS (G, H)

(arrows point to myofibroblasts; g–i

to endothelial cells). Corresponding

merged images (I, i); yellow indicates

colocalization. Staining for CD311 en-

dothelial cells and fibromodulin in nor-

mal skin, respectively (J, K) (arrows

point to blood vessels). Corresponding

merged image (L); yellow indicates co-

localization. Staining for CD311 endo-

thelial cells and fibromodulin in HTS,

respectively (M, N) (arrows point to en-

dothelial cells; arrowheads indicate en-

dothelial cells around blood vessels).

Corresponding merged image (O).

HTS, hypertrophic scar; a-SMA; a-

smooth muscle actin. Scale bar: 50mm.



Fibroblasts, which play a critical role in fibrillogenesisand wound healing, may be responsible in the develop-ment of HTS and excessive amounts of collagen synthe-sized during wound healing. Normal skin fibroblastshad increased staining intensity for decorin (Figure 6A)as compared with HTS fibroblasts (Figure 6B); whereas,biglycan immunoreactivity was lower in normal skin fibro-blasts (Figure 6D) compared with HTS cells (Figure 6E).Fibromodulin was also lower in fibroblasts cultured fromHTS (Figure 6H) compared with normal skin fibroblasts(Figure 6G). No obvious change was observed for lumicanimmunoreactivity between fibroblasts from normal skinand HTS tissue (Figure 6J and K, respectively). To inves-tigate the differences in the protein expression of SLRPsbetween fibroblasts from normal skin and HTS, cells were

isolated and cultured in DMEM supplemented with10% FBS or for 24 hours in 10% FBS/DMEM followingincubation in serum-free DMEM until they were 95%confluent. Both experimental settings yielded similar re-sults; however, cells grown in media supplemented contin-uously with FBS showed more obvious changes andclear immunoblotting bands. Figure 6 shows represen-tative data from fibroblasts cultured continuously inDMEM with 10% FBS. The protein expression level ofdecorin and fibromodulin were significantly down-regu-lated in HTS, while biglycan was higher when comparedwith fibroblasts from normal skin (Figure 6C, F, and I).Lumican protein expression level did not show signifi-cant differences between normal and HTS fibroblasts(Figure 6L).

Figure 6. Expression of SLRPs in nor-

mal skin and HTS-cultured fibroblasts.

Immunofluorescence staining for

decorin, biglycan, fibromodulin, and

lumican, respectively, in cultured fibro-

blasts from normal uninjured skin (A,

D, G, J). Immunofluorescence staining

for decorin, biglycan, fibromodulin, and

lumican, respectively, in HTS-cultured

fibroblasts (B, E, H, K). Western blot

analysis of decorin, biglycan, fibro-

modulin, and lumican, respectively, of

proteins extracted from cultured

fibroblasts (C, F, I, L). N, normal skin

fibroblasts; SLRPs, small leucine-rich

proteoglycans; HTS, hypertrophic scar

fibroblasts; F. modulin, fibromodulin;

Actin represents loading control. Scale

bar: 50mm.



DISCUSSION

Despite an increased understanding of the biological pro-cesses that result in fibrosis and scar formation, detailedevidence is still required to elucidate factors that lead tothe development of HTS. Studying the molecular and cel-lular mechanisms involved in tissue repair and woundhealing may provide information that could potentiallyhelp clinicians develop therapeutic strategies that minimizefibrosis or reduce HTS formation following thermal, sur-gical, or traumatic injuries. Because the focus of this workis to establish optimal wound healing conditions andmethodology that mimics normal dermal regeneration,differences in the localization, accumulation and expres-sion of the SLRPs, decorin, biglycan, fibromodulin, andlumican were investigated in HTS and compared with nor-mal unwounded skin. Our previous studies5,14 showed thatin mature postburn scars the immunostaining of decorinand biglycan was intermediate between normal dermis andthat in the HTS indicating the changes observed in currentstudy seen in HTS are greater than those can be detected inminimal scarring. The data suggest that down-regulationof the SLRPs expression may coincide with aggravation ofthe scar formation. The subjects of this study were selectedfrom a broad range of burn patients in terms of gender,age, the extent of injury, and the length of postwounding.Interestingly, we discovered the same trend of changes inSLRPs expression in all subjects despite demographic het-erogeneity suggesting that the findings may be true repre-sentation of changes in hypertrophic scarring and/or otherfibroproliferative disorders. The major finding was the ex-pression of decorin and fibromodulin were significantlylower in HTS tissue compared with normal skin.

Reduced synthesis of decorin has been linked to the de-velopment of dermal scars while, a gradual increase in thelevel of decorin appears to stabilize wound healing and re-duce scar tissue at later time points.15 It is known that deepdermal injuries result in more pronounced scarring. Inter-estingly, compared with papillary and superficial dermis,there was an increased accumulation and thicker collagenfibril deposition in the deep dermis that coincided with alower expression of decorin in these areas. Lack of decorincan result in irregularity in the collagen fibers diameter andstructure and fragile skin with markedly reduced tensilestrength.16 In addition, given the role of decorin in the in-duction of apoptosis,17 a lack of decorin may contribute tothe development of hypercellularity in granulation tissue.Decorin knock-out mice showed a two-fold increase in thenumber of fibroblasts in their periodontal ligaments com-pared with wild-type animals.18 Hence, during dermalwound healing, down-regulation of decorin may result inan increased number of cells particularly, collagen-pro-ducing myofibroblasts that contribute to the developmentof HTS or tissue fibrosis.

Our findings also indicate that fibromodulin is signifi-cantly down-regulated in HTS when compared with nor-mal skin. Interestingly, fibromodulin overexpressionimproved tissue repair and reduced scarring in a modelof wound healing in rabbits.11 In a nonscarring modelof human gingival wound healing, fibromodulin showed a10-fold increase at later time points.4 In addition, an in-verse relationship between fibromodulin expression andscar formation in both fetal and adult rat wound healing9

suggests a role for fibromodulin in the reduction of scar-ring. It has been shown that fibromodulin decreased theexpression of TGF-b1 and TGF-b2, which was associatedwith an increase in the expression of antifibrotic TGF-b3,suggesting that fibromodulin may also be a key mediatorin preventing scar formation.11 Although Western blotanalysis did not show notable changes in the expression ofbiglycan and lumican between HTS tissue and normalskin, their mRNA expression levels were significantlyincreased suggesting a compensatory effect for decorin orfibromodulin reduction. Previously it has been reportedthat down-regulation of one SLRP leads to an increasedexpression of another SLRP of the same subclass.19,20

Taken together, the findings suggest that decreased expres-sion of fibromodulin and decorin at the same time duringskin wound repair may aggravate HTS formation and itsabnormal physical properties.

During wound healing, regeneration of the epidermisoccurs by reepithelialization via keratinocyte activationand migration. Faster reepithelialization and woundclosure results in improved wound healing and reducedscarring. The basal epithelium supports keratinocytes thatgive rise to the upper layers of the epidermis by differenti-ation. Spatiotemporal expression over time of biglycan,fibromodulin, and lumican in migrating and woundepithelium in gingival wound healing suggests that SLRPsare involved in reepithelialization.4 The role of SLRPs inthe reconstruction of epidermis following dermal injuriesis not clear; however, distinct localization of biglycan,fibromodulin, and lumican in the basement membraneand basal epithelial cells in normal unwounded dermissuggests a novel function for these SLRPs in the develop-ment of HTS.

Angiogenesis is a critical step in tissue growth and de-velopment and wound healing. Adult wounds that developscarring show an increased rate of angiogenesis, whereas,there is a marked decrease in neovascularization in fetalwounds.21,22 In these experiments there was a three- tofour-fold increase in the number of blood vessels in HTScompared with normal skin. It is not clear whether a lackof SLRPs may lead to an increased number of blood ves-sels in HTS and the role of SLRPs for angiogenesis in HTSwarrants further study. Neovascularization was not sig-nificantly changed in fibromodulin- or biglycan-deficientanimals but decorin deficiency leads to impaired angiogen-esis in injured mouse cornea.23 However, the inhibition ofECmigration by decorin may influence the assembly of thecell-associated ECM24 and impact the neovascularizationrate. Under certain conditions lumican may slow down theformation of blood vessels, which is supported by evidenceshowing that lumican inhibits angiogenesis in mice.25

Fibroblasts play a key role in wound healing, cytokinerelease, ECM production, and collagen turnover. In addi-tion to earlier findings for decorin in HTS,5,14,26 the currentstudy found the expression of fibromodulin was also signifi-cantly lower in cultured HTS fibroblasts than in normal skincells. Further investigation is needed to elucidate the cellularand molecular mechanisms that lead to the reduction ofdecorin and fibromodulin expression in HTS tissue. Differ-ences in TGF-b expression by fibroblast subtypes in HTSand within normal skin may play an important role. Forexample, HTS may occur following injuries that affect thedeeper layers of the dermis which contains larger numbers of



reticular fibroblasts which overexpress TGF-b resulting inHTS formation.26 In contrast, superficial wounds involvingpapillary fibroblasts tend to heal with minimal scarring. Inaddition to binding to ECM molecules and growth factors,SLRPs also interact with cells. Association of SLRPs withfibroblasts3,4 and capillary EC27 suggests that SLRPs alsointeract with these cells to modulate their function.

The results described above warrant further studies tounderstand changes in the SLRPs expression patternfollowing injury and HTS formation. Changes in theglycosylation pattern of SLRPs can influence the develop-ment of HTS and may be the future direction of this study.Changes in proteoglycan glycosylation alter proteoglycancontent in soft tissues and correlates with modifications inthe ECM organization and thereby alter collagen fibrillo-genesis. For example, changes in the number of N-linkedoligosaccharides attached to the core protein of decorin28

affect decorin degradation states and hence influence ma-trix remodeling and the activity of TGF-b. Findings revealthat decorin is glycosylated with longer galactosamino-glycan side chains in smooth muscle neoplasm and led tochanges in the tissue architecture and ECM remodeling.29

In addition, scarring appeared to result in a reduction inhyaluronan and CS/DS.30

Taken together, the mechanisms that lead to the reduc-tion of scarring by SLRPs still need considerable investiga-tion. However, at least two mechanisms appear to besignificant for the SLRPs antiscarring properties. First,SLRPs can bind to type I collagen to delay fibrillogenesisand induce formation of densely organized packed and thin-ner fibrils in vivo.31–33 This is in contrast to collagen archi-tecture in HTS that shows thick, unorganized, and looselypacked collage fiber bundles. Second, it has been shown thatdecorin, biglycan, and fibromodulin can bind to TGF-b1through their core proteins and inhibit TGF-b1 fibrogenicactivity by preventing the interaction of TGF-b with its cellsurface receptors.8–10 Therefore, the presence of decorinmaydown-regulate TGF-b expression and/or activity duringwound healing leading to reduced scar formation.

Thus, SLRPs are important ECM molecules that regu-late wound healing by regulating collagen fibrillogenesisand turnover as well as modulating growth factor activityand interacting with different connective tissue cells. Thisstudy points out the differences in the localization andexpression of SLRPs in HTS. The role of SLRPs in mod-ulating cell functions during wound healing, tissue regen-eration and the development of HTS warrants furtherfunctional studies. A better understanding of cellularand molecular mechanisms involved in dermal woundhealing may provide valuable information that is usefulto prevent and treat HTS and other forms of fibroprolifer-ative disorders.

ACKNOWLEDGMENTS

The wound healing research in the author’s laboratory isfunded by the Canadian Institutes of Health Research, theAlberta Heritage Foundation for Medical Research andFirefighters’ Burn Trust Fund of the University of Alberta.

This work was supported by the Firefighters’ BurnTrust Fund of the University of Alberta Hospital and theCanadian Institutes of Heath Research.

REFERENCES

1. Iozzo RV. The biology of the small leucine-rich proteogly-

cans. Functional network of interactive proteins. J Biol Chem

1999; 274: 18843–6.2. Kresse H, Schonherr E. Proteoglycans of the extracellular

matrix and growth control. J Cell Physiol 2001; 189: 266–74.3. Alimohamad H, Habijanac T, Larjava H, Hakkinen L. Co-

localization of the collagen-binding proteoglycans decorin,

biglycan, fibromodulin and lumican with different cells in

human gingiva. J Periodontal Res 2005; 40: 73–86.4. Honardoust D, Eslami A, Larjava H, Hakkinen L. Localiza-

tion of small leucine-rich proteoglycans and transforming

growth factor-beta in human oral mucosal wound healing.

Wound Repair Regen 2008; 16: 814–23.5. Scott PG, Dodd CM, Tredget EE, Ghahary A, Rahemtulla

F. Immunohistochemical localization of the proteoglycans

decorin, biglycan and versican and transforming growth

factor-beta in human post-burn hypertrophic and mature

scars. Histopathology 1995; 26: 423–31.6. Gallant CL, Olson ME, Hart DA. Molecular, histologic and

gross phenotype of skin wound healing in red Duroc pigs re-

veals an abnormal healing phenotype of hypercontracted,

hyperpigmented scarring. Wound Repair Regen 2004; 12:

305–19.7. Sappino AP, Masouye I, Saurat JH, Gabbiani G. Smooth

muscle differentiation in scleroderma fibroblastic cells. Am J

Pathol 1990a; 137: 585–91.8. Hildebrand A, Romaris M, Rasmussen LM, Heinegard D,

Twardzik DR, Border WA, Ruoslahti E. Interaction of the

small interstitial proteoglycans biglycan, decorin and fib-

romodulin with transforming growth factor beta. Biochem J

1994; 302: 527–34.9. Soo C, Hu FY, Zhang X, Wang Y, Beanes SR, Kim SJ,

Longaker MT, Freymiller E, Ting K. Differential expression

of fibromodulin, a transforming growth factor-beta modula-

tor, in fetal skin development and scarless repair. Am J Pa-

thol 2000; 157: 423–33.10. Kolb M, Margetts PJ, Sime PJ, Gauldie J. Proteoglycans de-

corin and biglycan differentially modulate TGF-beta-medi-

ated fibrotic responses in the lung. Am J Physiol Lung Cell

Mol Physiol 2001; 280: 1327–34.11. Stoff A, Rivera AA, Mathis JM, Moore ST, Banerjee NS,

Everts M, Espinosa-de-los-Monteros A, Novak Z, Siegal

GP, Stoff-Khalili MA, Curiel DT. Effect of adenoviral

mediated overexpression of fibromodulin on human dermal

fibroblasts and scar formation in full-thickness incisional

wounds. J Mol Med 2007; 85: 481–96.12. Zhang Z, Li XJ, Liu Y, Zhang X, Li YY, Xu WS. Recombi-

nant human decorin inhibits cell proliferation and down-

regulates TGF-beta1 production in hypertrophic scar

fibroblasts. Burns 2007; 33: 634–41.13. Zhang Z, Liu Y, Zhang X, Xu WS. The content of decorin

and its mRNA expression in normal human skin and hyper-

plastic scars. Zhonghua Shao Shang Za Zhi 2004; 20: 76–8.14. Scott PG, Dodd CM, Ghahary A, Shen YJ, Tredget EE. Fi-

broblasts from post-burn hypertrophic scar tissue synthesize

less decorin than normal dermal fibroblasts. Clin Sci (Lond)

1998; 94: 541–7.15. Sayani K, Dodd CM, Nedelec B, Shen YJ, Ghahary A, Tred-

get EE, Scott PG. Delayed appearance of decorin in healing

burn scars. Histopathology 2000; 36: 262–72.



16. Danielson KG, Baribault H, Holmes DF, GrahamH, KadlerKE, Iozzo RV. Targeted disruption of decorin leads to ab-normal collagen fibril morphology and skin fragility. J CellBiol 1997; 136: 729–43.

17. Stadler M, Naumann U, Wick W, Weller M. Transforminggrowth factor-beta and p-21: multiple molecular targets ofdecorin-mediated suppression of neoplastic growth. Cell Tis-sue Res 1999; 296: 221–7.

18. Hakkinen L, Strassburger S, Kahari VM, Scott PG, Eich-stetter I, Lozzo RV, Larjava H. A role for decorin in thestructural organization of periodontal ligament. Lab Invest2000b; 80: 1869–80.

19. Svensson L, Aszodi A, Reinholt FP, Fassler R, Heinegard D,Oldberg A. Fibromodulin-null mice have abnormal collagenfibrils, tissue organization and altered lumican deposition intendon. J Biol Chem 1999; 274: 9636–47.

20. Zhang G, Ezura Y, Chervoneva I, Robinson PS, Beason DP,Carine ET, Soslowsky LJ, Iozzo RV, Birk DE. Decorinregulates assembly of collagen fibrils and acquisition ofbiomechanical properties during tendon development. J CellBiochem 2006; 15: 1436–49.

21. Whitby DJ, Ferguson MW. Immunohistochemical localiza-tion of growth factors in fetal wound healing. Dev Biol 1991;147: 207–15.

22. Inhara S, Motobayashi Y, Nagao E, Kistler A. Ontogenetictransition of wound healing pattern in rat skin occurring atthe fetal stage. Development 1990; 110: 671–80.

23. Schonher E, Sunderkotter C, Schaefer L, Thanos S, GrasselS, Oldberg A, Iozzo RV, Young MF, Kresse H. Decorindeficiency leads to impaired angiogenesis in injured mousecornea. J Vasc Res 2004; 41: 499–508.

24. Kinsella MG, Fischer JW, Mason DP, Wight TN.Retrovirally mediated expression of decorin by macro-vascular endothelial cells. Effects on cellular migration andfibronectin fibrillogenesis in vitro. J Biol Chem 2000; 275:13924–32.

25. Albig AR, Roy TG, Becenti DJ, Schiemann WP. Transcrip-tome analysis of endothelial cell gene expression inducedby growth on matrigel matrices: identification and character-ization of MAGP-2 and lumican as novel regulators ofangiogenesis. Angiogenesis 2007; 10: 197–216.

26. Wang J, Dodd C, Shankowsky HA, Scott PG, Tredget EE.Deep dermal fibroblasts contribute to hypertrophicscarring. Wound Healing Research Group. Lab Invest 2008;88: 1278–90.

27. Jarvelainen H, Puolakkainen P, Iozzo RV, Sage EH, WightTN. A role for decorin in cutaneous wound healing andangiogenesis. Wound Repair Regen 2006; 14: 443–52.

28. Yoon JH, Brooks RL Jr., Zhao JZ, Isaacs D, Halper J.The effects of enrofloxacin on decorin and glycosaminogly-cans in avian tendon cell cultures. Arch Toxicol 2004; 78:599–608.

29. Berto AG, Sampaio LO, Franco CR, Cesar RM Jr., Mi-chelacci YM. A comparative analysis of structure and spatialdistribution of decorin in human leiomyoma and normalmyometrium. Biochim Biophys Acta 2003; 1619: 98–112.

30. Hahn MS, Jao CY, Faquin W, Grande-Allen KJ.Glycosaminoglycan composition of the vocal fold laminapropria in relation to function. Ann Otol Rhinol Laryngol2008; 117: 371–81.

31. Hedbom E, Heinegard D. Binding of fibromodulin anddecorin to separate sites on fibrillar collagens. J Biol Chem1993; 268: 27307–12.

32. Font B, Eichenberger D, Goldschmidt D, Boutillon MM,Hulmes DJ. Structural requirements for fibromodulin bind-ing to collagen and the control of type I collagen fibrillogen-esis critical roles for disulphide bonding and the C-terminalregion. Eur J Biochem 1998; 254: 580–7.

33. Vogel KG, Koob TJ, Fisher LW. Characterization andinteractions of a fragment of the core protein of the smallproteoglycan (PGII) from bovine tendon. Biochem BiophysRes Commun 1987; 148: 658–63.



Small leucine-rich proteoglycans, decorin and fibromodulin, are reduced in postburn hypertrophic...

Documents

Transcript of Small leucine-rich proteoglycans, decorin and fibromodulin, are reduced in postburn hypertrophic...