Lumican Accelerates Wound Healing by Enhancing α2β1 Integrin-Mediated Fibroblast Contractility

Lumican is a dermatan sulfate proteoglycan highly expressed in connective tissue and has the ability to regulate collagen fibril assembly. Previous studies have shown that lumican is involved in wound healing, but the precise effects of lumican on reepithelialization and wound contraction, the two pivotal aspects of skin wound healing, have not been investigated. Here we explored the roles of lumican in fibroblast contractility, a main aspect of skin wound healing, by adopting mice skin wound healing model and the corresponding in vitro cellular experiments. Our results showed that lumican can promote skin wound healing by facilitating wound fibroblast activation and contraction but not by promoting keratinocyte proliferation and migration. Silencing of integrin α2 completely abolished the pro-contractility of lumican, indicating lumican enhances fibroblast contractility via integrin α2. Our study for the first time demonstrated that lumican can affect fibroblast’s mechanical property, which is pivotal for many important pathological processes, such as wound healing, fibrosis, and tumor development, suggesting that lumican might have a potential to be used to modulate these processes.


Introduction
Wound healing is a complex and dynamic pathological process, finely regulated by intracellular signaling and extracellular components, including cytokines, growth factors, extracellular matrix (ECM) proteins, and so on [1,2].
Accumulating evidences have shown that ECM proteins play important roles in skin cells proliferation, differentiation or motility during wound healing process. Knockdown of osteopontin promotes wound repair and reduces scarring by reducing leukocyte recruitment and altering ECM deposition and angiogenesis [3]. Wound healing is delayed in sydecan-4 null mice, which is closely associated with reduced granulation tissue and wound contraction in sydecan-4 null mice compared to that in wild type littermates [4]. Dermatopontin, a tyrosine-rich ECM protein with a striking tendency to bind the small dermatan sulfate proteoglycan, can promote wound healing by affecting migration and proliferation of local cells [5].
Lumican, a small dermatan sulfate proteoglycan, is highly expressed in interstitial tissue [6] and can modulate collagen fibril organization [7], inflammatory cells migration, infiltration [8,9] and keratinocyte phenotypes [7,8,9,10]. Nevertheless, there is no report about the effects of lumican on fibroblast, one of the major cell types in wound healing. Corneal epithelial wound healing was retarded by knockout of lumican or anti-lumican antibody [11], while it was promoted by purified lumican protein [10]. Further investigation revealed that lumican is also required for skin wound repair by showing that skin wound healing is significantly delayed in lumican null mice compared to that in wild type mice [7]. Although the importance of lumican in wound healing had been established, the mechanisms of how lumican affects wound healing, particularly fibroblast behaviors, has not been studied yet.
In this study, we investigated the underlying mechanisms of how lumican affects wound healing. We first applied purified lumican protein to the mice skin wound model and found that lumican promoted mice skin wound healing. Second, we revealed that recombinant lumican enhanced fibroblasts activation but not epidermis migration or skin cells proliferation in vivo by using immunohistochemistry analyses, and these results are consistent with our in vitro cellular function assays, which showed that recombinant lumican enhanced the contractility of fibroblasts in three-dimensional surroundings, but neither the migration of keratinocytes nor the proliferation of these two cell types. Further analyses revealed that silencing of integrin a2 but not integrin a1 inhibited the contractility promoting ability of lumican, indicating that effects of lumican on fibroblast contractility in wound healing is mainly mediated by integrin a2.

Lumican Recombinant Protein and Lumican Core Protein Expression, Purification and Characterization
Lumican ORF was cloned into the episomal expression vector V152 with pCEP-Pu-Strep II-tag (C-terminal) in-frame and the sequence of the BM-40 (SPARC/osteonectin) signal peptide downstream of the CMV promoter. Lumican was recombinant expressed in EBNA-293 cells after transfecting reconstructed plasmid by using X-tremeGENE 9 DNA Transfection Reagent (Roche, Mannheim, Germany). 2 days after transfection, the cells were screening with 2 mg/ml puromycin (Sigma-Aldrich, St. Louis, MO) in Dulbecco's modified Eagle's medium (DMEM, 11965, Gibco, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS, Gibco-BRL, Gaithersburg, MD) for 7 days, then the culture media was collected and applied to the Strep Tactin sepharose column (IBA, Gottingen, Germany). After this, the column was washed with binding buffer and eluted by elution buffer containing 2.5 mM desthiobiotin. The collected fractions were further quantified by a Nanodrop 2000 spectrophotometer (Thermo Fisher Scientific, Wilmington, DE) and BCA Protein Assay Kit (Pierce. Biotechnology Inc, Rockford, IL) and identified by western blot assay.
Lumican core protein was expressed and purified as previously described [12]. The collected core protein was identified by western blot assay, the concentration of core protein was measured by a Nanodrop 2000 spectrophotometer and BCA Protein Assay Kit. The purity of lumican recombinant protein and lumican core protein was determined by Coomassie brilliant blue staining.

Deglycosylation of Lumican Recombinant Protein
Deglycosylation of lumican recombinant protein was conducted by using N-glycosidase F (New England Biolabs, Beverly, MA) according to the manufacturer's instructions.

Mice Wound Healing Model
Six-week-old male Balb/c mice were purchased from the Shanghai Slac Laboratory Animal CO. LTD (Slac, Shanghai, China). Mice wound healing experiment was conducted in Shanghai Jiao Tong University. All procedures were approved by the Shanghai Jiao Tong University Animal Care and Use Committee; all animals were housed under standard conditions under institution-approved guidelines.
Briefly, after anesthesia, the dorsal hair of mice was shaved, and a 5 mm round full-thickness skin excision wound was made at midline of the back 2-cm caudal ward from the skull of each mouse. Lumican recombinant protein (20 nM) or adjuvant (elution buffer with 30% glycerol) was topical applied to wound from post-injury day 0 to day 7 once daily. 30 mice were divided into two groups, one group was treated with lumican, and the other group was treated with adjuvant as control. Wound area was measured at post-injury day 0, 3, 7, 10 and 14. At post-injury day 3, 7 and 14, wound tissues of 5 mice of each group were harvested for further analyses. 10 mice were used to investigated the roles of lumican core protein in dorsal skin wound healing. Lumican core protein (20 nM) or adjuvant (elution buffer with 30% glycerol) was applied to wound from post-injury day 0 to day 7 once daily, Wound area was measured at post-injury day 0, 3, 7, 10 and 14.

Hematoxylin and Eosin (HE) and Immunohistochemistry Stain
Paraformaldehyde-fixed, paraffin-embedded tissue sections (5 mm) at the midlines of wound specimen were adopted for HE and immunohistochemistry stain. For immunohistochemistry stain, the sections were detected with primary antibodies against proliferating cell nuclear antigen (PCNA, Epitomics, Burlingame, CA), a-smooth muscle actin (a-SMA, Sigma-Aldrich, St Louis, MO) overnight at 4uC. After incubated with appropriate secondary antibodies, the sections were developed with diaminobenzidine and counterstained with haematoxylin. Images were processed for publication using Adobe Photoshop CS4.

Cell Culture
Immortal human keratinocytes (HaCaT) (China Center for Type Culture Collection, Wuhan, China) [13], primary human dermal fibroblasts (HDF) and human embryonic kidney 293/ Epstein-Barr nuclear antigen cells (EBNA-293) were maintained in DMEM supplemented with 10% FBS, 100 U/ml penicillin and 100 mg/ml streptomycin (Solarbio, Beijing, China). HDF cells were obtained from adolescent foreskin tissue of patients of Shanghai Jiao Tong University affiliated Sixth People's Hospital with written informed consent signed by their parents or guardians on the behalf of the participants according to Declaration of Helsinki Principles. All cellular experiments were conducted in Shanghai Jiao Tong University. The study was approved by the Ethics Committee of Shanghai Jiao Tong University affiliated Sixth People's Hospital. All cells were incubated at 37uC humidified atmosphere with 5% CO2. HDF at passage 4-8 were adopted for experiments.

Cell Proliferation Assay
Proliferation was determined by using a standard CCK-8 cell counting kit (Dojindo, Kumamoto, Japan) according to the manufacturer's instruction. HaCaT (2.0610 4 cells/ml) and HDF cells (4.0610 4 cells/ml) were seeded in 96-well plates (100 ml/ well). The medium was replaced 24 hours later, and lumican recombinant protein (0, 0.1 nM, 1 nM, 10 nM) was added in. The medium was changed once every two days and optical density (OD 450 ) values were measured at day 0, 2, 4 and 6 respectively by a microplate reader (SpectraMax M5, Molecular device).

Scratch Wound Healing Assay
HaCaT cells at 3610 5 /ml were seeded in 24-well plates (500 ml/well) and cultured to confluence as a monolayer, and starved by DMEM supplied with 1% FBS for 24 hours, then gently scratch the monolayer across the centre of the cell with a 10 ml pipette tip. Wash with PBS for 3 times and changed by DMEM supplied with 1% FBS and different concentration of lumican recombinant protein (0, 0.1 nM, 1 nM, 10 nM). The gaps were photographed every 6 hours until 24 hours post-scratch by a live cell imaging system (Olympus, Tokyo, Japan).
Fibroblast Activation Assay 1610 4 /well HDF cells were seeded in 24-well plates with serum-free DMEM medium, and lumican recombinant protein (10 nM) or adjuvant control was added in to treat for 3 days. Then cells were immunofluorescence stained as described [14]. DAPI (Sigma-Aldrich, St Louis, MO) was used to stain cell nucleus, and a-SMA primary antibody (Sigma-Aldrich, St Louis, MO) was used to stain a-SMA. The percentage of activated fibroblasts were calculated as a-SMA positive cells divided DAPI positive cells in 5 random fields of each well.

3D Collagen Gel Contraction Assay
3D collagen gel contraction assays were conducted as reported [14]. Lumican recombinant protein at 0, 0.1 nM, 1 nM or 10 nM was added into the collagen gel lattices.
For activated myofibroblasts gel contraction assay, fibroblasts were activated by TGF-b1 (2 ng/ml) for 5 days, then collected and used.

SiRNA Transfection
SiRNA duplexes targeted at integrin a1 or a2 and scramble control siRNA duplex were obtained from GenePharma (Shanghai, China). Transfection was performed by using the Lipofectamine RNAiMAX Reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instruction.

Western Blot Assay
The primary antibodies against integrin a1 (abcam, Cambridge, MA), integrin a2 (abcam, Cambridge, MA) and GAPDH (abcam, Cambridge, MA) were incubated overnight at 4uC. Signals on the membranes were detected by an Odyssey infrared imaging system (LI-COR, Lincoln, NE) after incubating with the IRDye 800 antirabbit or IRDye 680 anti-mouse (LI-COR, Lincoln, NE) secondary antibodies for 1 hour at room temperature.

Statistical Analysis
Statistic differences were calculated using two-tailed Student's t test. p,0.05 was considered statistically significant.

Lumican and Lumican Core Protein Promotes Mice Skin Wound Healing
To investigate the in vivo roles of lumican in skin wound healing as a secreted protein, we purified recombinant lumican protein and lumican core protein, then applied them to a mice fullthickness skin excision wound model. After 7 days of daily topical administration, we found that recombinant glycosylated lumican protein and lumican core protein can both significantly promoted mice skin wound healing. As illustrated in figure 1A, lumican recombinant protein expressed by EBNA-293 cells is glycosylated, and can be deglycosylated by N-glycosidase F [12]. The wound area of lumican or lumican core protein-treated mice was obviously smaller than that of adjuvant-treated mice from postinjury day 3 to day 14 ( Figure 1B-E). Because the adjuvant for lumican recombinant protein and lumican core protein is different, we didn't compare the efficiency between these two kind of proteins. The general condition was not different between these groups of mice (data not shown).
The result indicates that the wound healing promoting effect of lumican is at least not totally dependent on the dermatan sulfate chain.

Recombinant Lumican Protein does not Significantly Affect Activation of Keratinocytes
Skin wound healing is coordinately accomplished by two major events, wound contraction and reepithelialization [15]. Wound contraction is primarily generated by fibroblast activation and contraction, while reepithelialization is a process constructed by migration and proliferation of keratinocyte [16], which is also termed as keratinocyte activation [17,18].
To explore the mechanism of how lumican affects wound healing, we first evaluated the roles of recombinant lumican protein in keratinocyte proliferation in vivo and in vitro. No obvious difference was observed on PCNA staining in wound epidermis between lumican-treated mice and adjuvant-treated mice (Figure 2A and 2B). And in vitro CCK-8 proliferation assay also displayed that recombinant lumican protein did not affect HaCaT cells proliferation ( Figure 2C).
Then in vivo migration of epidermal cells was evaluated by measuring the length of epidermal tongue in wound sections. The results showed that lumican had no effect in the length of epidermal tongue in vivo ( Figure 3A and 3B). Scratch wound healing assays were also conducted to examine the effect of lumican on HaCaT cell migration in vitro, the results displayed no significant difference between lumican-treated and adjuvanttreated cells ( Figure 3C).
Together, these data indicate that recombinant lumican protein has no effect on keratinocyte activation both in vivo and in vitro.

Recombinant Lumican Protein Promotes Fibroblast Activation in vivo and in vitro
We further examined the effect of lumican on fibroblast behaviors in vivo and in vitro. a-SMA, a fibroblast activation marker, was immunostained in lumican-treated or adjuvant treated mice wound sections. The results showed that the ratio of a-SMA positive stromal cells in wound granulation tissue of lumican-treated mice was significantly higher than that of adjuvant-treated mice at 3 and 7 days post-injury, but no difference was observed at 14 days post-injury ( Figure 4A-4C).
The results indicate that topical usage of recombinant lumican protein can promote fibroblast activation in the early wound healing process, while this effect would not last to later proliferation stage when activated fibroblasts would contribute to excessive scar formation.
We also investigated the in vitro effects of lumican on fibroblasts activation. As displayed in figure 4D and 4E, the ratio of a-SMA positive cells of lumican treated HDF was significantly higher than that of adjuvant treated HDF. That means lumican can also activate HDF cells in vitro.
Besides fibroblast activation, fibroblast proliferation is also involved in wound repair. We further investigated the effects of lumican on fibroblast proliferation in vivo and in vitro. Evaluated by PCNA staining, the stromal cell proliferation in wound granulation tissue was similar between lumican-treated mice and adjuvant-treated mice at 3 and 7 days post-injury ( Figure 5A and 5B). And in vitro CCK-8 proliferation assay also showed that lumican did not promote or inhibit fibroblast proliferation ( Figure 5C).

Lumican Enhances Fibroblast Contractility in Threedimensional Surroundings
Because activated fibroblast is defined not only by a-SMA expression, but also by higher contractility [19]. To examine the effect of lumican on fibroblast contractility, three-dimensional collagen gel contraction assay was performed. The results showed that lumican can enhance contractility of fibroblast in a dosedependent manner, further confirming our in vivo finding that lumican can promote fibroblast activation ( Figure 6A and 6B).
Besides in quiescent fibroblasts, we also tested the procontractile ability of lumican on TGF-b1 activated myofibroblasts. After activation with TGF-b1 (2 ng/ml) for 5 days, HDF cells were collected and used. The result showed that lumican can also enhance the contractile ability of activated fibroblasts ( Figure 6C and 6D), and the pro-contractility effect of lumican in quiescent fibroblasts and activated fibroblasts is not significantly different ( Figure 6E).
Taken together, these results demonstrate that recombinant lumican protein facilitates wound healing by promoting fibroblast activation but not by accelerating wound reepithelialization.

Silencing of Integrin a2 Blocks the Pro-contractility Effect of Recombinant Lumican Protein
Fibroblast contraction requires participation of ECM, focal adhesion and cytoskeleton [20], while integrins are the core of focal adhesion, which mediate the connection between ECM and cytoskeleton [21,22]. Integrin a1 and a2 are two highly expressed integrins in fibroblasts that recognize collagen [23]. Previous studies had revealed that lumican can interact with integrin a2 in endothelial cells [24] and melanoma cells [25]. So we presumed that lumican might enhance fibroblast contractility via integrin a1b1 or a2b1.
To verify this hypothesis, we conducted fibroblast collgen lattice contraction assay after silencing of integrin a1 or a2. The results showed that silencing of integrin a1 didn't block the procontractility of lumican in three-dimensional condition ( Figure 7A-C), but silencing of integrin a2 completely abolished the pro-contractility of lumican ( Figure 8A-C), clearly indicates that the pro-contractile effect of lumican in fibroblast is mainly dependent on integrin a2.

Discussion
Lumican is a highly-expressed secreted protein in connective tissue and has the ability to regulate collagen fibril organization [7]. It has been demonstrated that lumican is involved in inflammatory cells infiltration [7] and angiogenesis [24], the two substantial aspects of wound healing. Lumican was also suggested to be involved in reepithelialization in corneal wound model. However, the roles of lumican in wound contraction and epidermal coverage have not been explored in skin wound yet. Our study explored these two aspects and demonstrated that lumican can facilitate skin wound healing by enhancing fibroblast contractility via integrin a2, which deepens our understanding about the roles of lumican in interstitial tissue by displaying that lumican can modulate activation of fibroblast, the major stromal cell in connective tissue.
To the best of our knowledge, this is the first report about the roles of lumican on fibroblast behaviors. Fibroblast plays a fundamental role in wound healing by secreting chemokines, growth factors and ECM proteins that is necessary for inflammatory cells infiltration and neoangiogenesis [15]. Activated fibroblasts express a-SMA, which upregulates the contractility of fibroblasts [19], thus promote wound contraction by exerting higher contractile force to wound edges [26,27,28]. Wound contraction can shrink the size of wound, thus facilitates wound repair, so it is a crucial event in wound healing process [29]. And 3D collagen gel contraction assay simulates the 3D collagen-riched environment of wound granulation tissue, so it is a useful in vitro model for studying wound contraction [30]. It is difficult to quantitatively measure wound contraction in vivo. That's why we didn't observe wound contraction in mice directly, but our 3D collagen gel contraction assay alternatively proved that lumican can promote wound contraction.
Previous studies have shown that lumican exerts its functions predominantly through integrins [8,24,25,31,32]. But lumicanintegrin interactions are different among different cell types. As reported, lumican promotes neutrophil migration by interaction with integrin b2 [8], while it inhibits melanoma cell migration through integrin a2b1 [25,32]. In human umbilical vein endothelial cells, lumican blocks a2b1 integrin-mediated proangiogenic activity [24]. In this study we clearly demonstrated that lumican promotes dermal fibroblast contractility via integrin a2. The diversity of lumican functions through interaction with different integrins in different physiological and pathological conditions needs to be further investigated.
Tissue stiffness is implicated in many crucial physiological and pathological conditions, such as development, tumorigenesis, wound healing and fibrosis [33,34,35]. Tissue stiffness is primarily decided by two factors, thus ECM rigidity and contractile force generated by stromal cells [36,37]. Lumican can bind with collagen fibril, the major ECM structure components, and regulate its assembly [38],which might affect ECM rigidity. Our studies also show that lumican can modulate mechanical properties of fibroblast, the primary stromal cell in skin, viscera organs [39] and many malignancy tumors [40], indicating that lumican might involve in regulation of mechanical aspects of microenvironment in these situations. So further exploration aiming at lumican's roles in such fields might be interesting and fruitful.
Wound healing is a disturbing clinical situation with high morbidity and mortality [41], and satisfactory topical used drugs are still lacking [42]. Here we demonstrated that topical use of recombinant lumican protein can effectively promote mice skin wound healing, suggeting that lumican might be a potential bioagent for human wound repair.