Hyaluronidase-2 Regulates RhoA Signaling, Myofibroblast Contractility, and Other Key Profibrotic Myofibroblast Functions

rown Copyright a 2020 Published by Elsev his is an open access article under the CC B ttps://doi.org/10.1016/j.ajpath.2020.02.012 Hyaluronidase (HYAL)-2 is a weak, acid-active, hyaluronan-degrading enzyme broadly expressed in somatic tissues. Aberrant HYAL2 expression is implicated in diverse pathology. However, a significant proportion of HYAL2 is enzymatically inactive; thus the mechanisms through which HYAL2 dysregulation influences pathobiology are unclear. Recently, nonenzymatic HYAL2 functions have been described, and nuclear HYAL2 has been shown to influence mRNA splicing to prevent myofibroblast differentiation. Myofibroblasts drive fibrosis, thereby promoting progressive tissue damage and leading to multimorbidity. This study identifies a novel HYAL2 cytoplasmic function in myofibroblasts that is unrelated to its enzymatic activity. In fibroblasts and myofibroblasts, HYAL2 interacts with the GTPase-signaling small molecule ras homolog family member A (RhoA). Transforming growth factor beta 1edriven fibroblast-to-myofibroblast differentiation promotes HYAL2 cytoplasmic relocalization to bind to the actin cytoskeleton. Cytoskeletal-bound HYAL2 functions as a key regulator of downstream RhoA signaling and influences profibrotic myofibroblast functions, including myosin light-chain kinase emediated myofibroblast contractility, myofibroblast migration, myofibroblast collagen/fibronectin deposition, as well as connective tissue growth factor and matrix metalloproteinase-2 expression. These data demonstrate that, in certain biological contexts, the nonenzymatic effects of HYAL2 are crucial in orchestrating RhoA signaling and downstream pathways that are important for full profibrotic myofibroblast functionality. In conjunction with previous data demonstrating the influence of HYAL2 on RNA splicing, these findings begin to explain the broad biological effects of HYAL2. (Am J Pathol 2020, 190: 1236e1255; https://doi.org/10.1016/j.ajpath.2020.02.012)

Hyaluronan (HA) is a linear glycosaminoglycan, which is a ubiquitous component of extracellular matrix and has a major role in regulating cellular processes, such as cellecell adhesion, 1 migration, 2e4 differentiation, 5,6 and proliferation. 7e9 HA is therefore implicated in influencing numerous biological processes, and dysregulation of HA synthesis, turnover, and binding interactions contributes to a multitude of disease states, such as atherosclerosis, chronic inflammation, cancer progression, and fibrosis. 10e13 Hyaluronidase (HYAL)-2 has been identified as one of the principal enzymes involved in HA catabolism in vertebrates. HYAL2 is broadly expressed in tissues but has catabolic function within only a narrow acidic pH range (optimal pH, 4), and compared to other HYALs, has only weak intrinsic HA-degrading activity. 14 HYAL2 was originally identified as a lysosomal enzyme but was also subsequently identified as anchored to the cell membrane via a glycosylphosphatidylinositol link. 15 Aberrant expression of HYAL2 is implicated in diverse pathology, including cardiac and skeletal abnormalities, hematopoietic and platelet dysfunction, cancer, and fibrosis. 16e21 However, many reports indicate that a significant proportion of expressed HYAL2 may be enzymatically inactive; thus the cellular function of HYAL2 and the mechanisms through which HYAL2 dysregulation influences pathology have been previously unclear. 14,15,22 A number of studies have identified that HYAL2 can also have important nonenzymatic functions: Glycosylphosphatidylinositol-anchored HYAL2 has been identified as acting as a co-receptor for the transmembrane glycoprotein CD44, as a regulator of transforming growth factor (TGF)-b1emediated intracellular WW domainecontaining oxidoreductase 1 signaling, and as a viral entry receptor. 23e26 More recently, it was determined that glycosylphosphatidylinositol-anchored HYAL2 can translocate to the nucleus and regulate alternative splicing events that influence differentiation to profibrotic cell phenotypes. 27 Myofibroblasts are the principal effector cells that drive progressive fibrosis, a process that underlies many organspecific diseases and contributes to the burden of multimorbid conditions, including chronic kidney disease, lung fibrosis, liver cirrhosis, and degenerative joint disease. 28e33 Therefore, the study of factors that can either promote or prevent cell differentiation to a myofibroblast phenotype is important in identifying new therapeutic approaches to the treatment of chronic disease. Myofibroblasts are derived from differentiation of resident fibroblasts, pericytes, or epithelial cells under the influence of circulating profibrotic cytokines, such as TGF-b1. 13,34e36 HA and the most widely expressed isoform of the HA receptor, CD44, are key mediators of myofibroblast differentiation. 37e42 Specifically, the presence of pericellular HA matrices tethered to cell-surface CD44 are essential for TGF-b1edriven myofibroblast differentiation. In contrast, cell-surface expression of an alternatively spliced variant isoform of CD44 (denoted CD44v7/8) promotes prevention and/or reversal of TGF-b1edriven myofibroblast differentiation by causing internalization of pericellular HA matrices. 27,43 Nuclear HYAL2 was identified as a key modulator of CD44 mRNA alternative splicing leading to attenuated standard CD44 expression, while augmenting CD44v7/8 splice variant expression.
The purpose of this study was to determine the function of HYAL2 in myofibroblasts relevant to its cell localization. We report that cytoplasmic HYAL2 has distinct functions compared to nuclear HYAL2. In contrast to our previous studies demonstrating the antifibrotic actions of nuclear HYAL2, 27 we show that cytoplasmic HYAL2 in myofibroblasts can bind to the actin cytoskeleton and function as a master regulator of TGF-b1edriven RhoA signaling. Through this, HYAL2 promotes key profibrotic myofibroblast functions, including myofibroblast contractility, collagen deposition, and CTGF (CCN2) and MMP2 mRNA expression. These emerging studies from our group and others begin to explain the broad biological effects of HYAL2 and identify it as a key molecule for future study in chronic disease.

Materials
All reagents were purchased from Sigma-Aldrich (Poole, UK) or Thermo Fisher Scientific (Paisley, UK) unless otherwise stated. Reverse-transcription reagents, siRNA transfection reagents, and real-time quantitative PCR (qPCR) primers and reagents were purchased from Thermo Fisher Scientific. Other reagents used were recombinant human TGF-b1 (R&D Systems, Abingdon, UK) and the RhoA inhibitor Rhosin (G04; Merck Millipore, Watford, UK).

Animal Experiments Using Ischemia Reperfusion InjuryeInduced Renal Fibrosis
Ten adult (8-weekeold to 12-weekeold) male Lewis rats weighing 180 to 220 g were used (Harlan Laboratories, Ltd., Derby, UK). The rats acclimated to their surroundings for 7 days, with housing, handling, and experimental procedures in accordance with the local institutional policies and procedures licensed by the UK Home Office under the Animals (Scientific Procedures) Act (1986). Rats (n Z 5 in each treatment group) were provided with analgesics (200 mg of buprenorphine dissolved in 500 mL of drinking water) from 24 hours before surgery until kidney retrieval. Animals were anesthetized with isoflurane, a midline laparotomy incision made, and the renal pedicles were identified and clamped for 45 minutes using a vascular clip (ischemia reperfusion injury group). The kidney was visually assessed for both ischemia upon clamping and reperfusion upon release of the clamp. Rats in the sham group underwent the same operation without renal pedicle clamping (n Z 5). The rats were maintained for 28 days in accordance with local institutional policies and procedures. At 28 days, kidney tissue was retrieved with the animals under terminal anesthesia and stored in formalin. Kidneys were later embedded in paraffin and sections of 4 mm in thickness were cut. Sections were deparaffinized and rehydrated using xylene and reducing concentrations of ethanol. Antigen retrieval was performed in an autoclave using sodium citrate buffer with Tween. After blocking of nonspecific sites, sections were incubated with 6 mg/mL anti-HYAL2 (goat anti-human polyclonal species reactivity includes rat; Abcam, Cambridge, UK) and antieasmooth muscle actin (SMA) monoclonal antibody 1A4 (mouse anti-human species reactivity includes rat; Thermo Fisher Scientific). Fluorescence-labeled secondary antibodies used were donkey anti-goat (H þ L) Alexa Fluor 555 (Thermo Fisher Scientific) for HYAL2 and goat anti-mouse IgG (H þ L) Alexa Fluor 488 (Thermo Fisher Scientific) for a-SMA. DAPI was used for nuclear staining and sections were analyzed using laser scanning confocal microscopy.

Cell Culture
Human lung fibroblasts (AG02262) were purchased from Coriell Cell Repositories (Coriell Institute for Medical

HYAL2 Governs RhoA Driven Cell Responses
The American Journal of Pathologyajp.amjpathol.org Research, Camden, NJ). The cells were cultured in Dulbecco's modified Eagle's low-glucose medium and Ham's F-12 containing 5 mmol/L glucose, 2 mmol/L L-glutamine, 100 U/mL penicillin, and 100 mg/mL streptomycin, and supplemented with 10% fetal bovine serum (Biological Industries Ltd., Cumbernauld, UK). The cells were maintained at 37 C in a humidified incubator in an atmosphere of 5% CO 2 , and fresh growth medium was added to the cells every 3 days until the cells were ready for experimentation. The cells were incubated in serum-free medium for 48 hours before use in all experiments (growth arrest), and all experiments were performed under serum-free conditions unless otherwise stated. All experiments were undertaken using cells at passages 6 to 10.

RT-PCR and qPCR
qPCR was used to assess mRNA expression levels of a-SMA (ACTA2), hyaluronidase 2 (HYAL2), extra domain-Aefibronectin (EDA-FN), transforming growth factor beta 1 (TGFB1), [cellular communication network factor 2 (CCN2), also known as connective tissue growth factor (CTGF)], fibronectin (FN1), collagen I [collagen type I alpha 1 chain (COL1A1)] and collagen type I alpha 2 chain (COL1A2), and matrix metallopeptidase 2 (MMP2). Primers were commercially designed and purchased from Thermo Fisher Scientific (Table 1). The cells were grown in 35-mm dishes and washed with phosphate-buffered saline (PBS) before lysis with TRI Reagent and RNA purification according to the manufacturer's protocol. Reversetranscription used high-capacity cDNA reversetranscription kits, according to the manufacturer's protocols (Thermo Fisher Scientific). The kits use the random primer method for initiating cDNA synthesis. As a negative control, reverse-transcription was done with RNase-free, sterile H 2 O replacing the RNA sample. qPCR was done using the ViiA 7 Real-Time qPCR System (Thermo Fisher Scientific) in a final volume of 20 mL per sample, as follows: 1 mL of reverse-transcription product, 0.6 mL of target gene forward primer, 0.6 mL of target gene reverse primer, 10 mL of Power SYBR Green PCR Master Mix, and 7.8 mL of sterile RNase-free water. Amplification was done using a cycle of 95 C for 15 seconds and 60 C for 1 minute for 40 cycles, followed by a melt-curve stage at 95 C for 15 seconds, 60 C for 1 minute, and a final step of 95 C for 15 seconds. qPCR was simultaneously performed for GAPDH (primers and probe commercially designed and purchased from Thermo Fisher Scientific) as a standard reference gene. As a negative control, qPCR was performed with nucleasefree, sterile H 2 O replacing the cDNA sample. The comparative C t method was used for relative quantification of gene expression. The C t (threshold cycle where amplification is in the linear range of the amplification curve) for the standard reference gene (GAPDH ) was subtracted from the target gene C t to obtain the DC t . The mean DC t values for replicate samples were then calculated. The expression of the target gene in experimental samples relative to expression in control samples was then calculated using the following equation: 2 À[DCt(1)ÀDCt(2)] , where DC t (1) is the mean DC t calculated for the experimental samples, and DC t (2) is the mean DC t calculated for the control samples.

Immunocytochemistry
Cells were grown to 70% confluence in eight-well Permanox chamber slides. The culture medium was removed, and the cells washed with sterile PBS before fixation in 4% paraformaldehyde for 10 minutes at room temperature. After fixation, cells were permeabilized with 0.1% (v/v) Triton X-100 in PBS for 10 minutes at room temperature. Slides were blocked with 1% bovine serum albumin (BSA) for 1 hour before a further washing step with 0.1% (wt/v) BSA in PBS. Subsequently, the slides were incubated with the primary antibody diluted in 0.1% BSA and PBS for 2 hours at room temperature. After a further washing step, slides were incubated with Alexa Fluor 488econjugated and/or Alexa Fluor 594econjugated secondary antibodies for 1 hour at room temperature. Cell nuclei were stained with Hoechst solution. Cells were then mounted and analyzed by confocal and fluorescent microscopy. The following primary antibodies were used: mouse anti-human a-SMA (Sigma-Aldrich) and rabbit anti-human HYAL2 antibody (Atlas Antibodies, Sigma-Aldrich). The following secondary antibodies were used: goat anti-mouse Alexa Fluor 488, goat anti-rabbit Alexa Fluor 594 (InvitroGen/Thermo Fisher Scientific). For visualization of F-actin, fluorescein isothiocyanateeconjugated phalloidin toxin was used in place TaqMan assay gene ID Hs00426835_g1 ( siRNA Transfection Fibroblasts were transiently transfected with specific siRNA nucleotides (Thermo Fisher Scientific) targeting HYAL2. Transfection was done using Lipofectamine 2000 transfection reagent (InvitroGen/Thermo Fisher Scientific) in accordance with the manufacturer's protocol. Briefly, cells were grown to 50% to 60% confluence in antibiotic-free medium in either 35-mm dishes or eight-well Permanox chamber slides. Transfection reagent (2% v/v) was diluted in Opti-MEM reduced growth medium (Gibco/Thermo Fisher Scientific) and left to incubate for 5 minutes at room temperature. HYAL2 siRNA (siHYAL2) oligonucleotides were diluted in Opti-MEM reduced growth medium to achieve a final concentration of 30 nmol/L. The transfection reagent and siRNA mixtures were then combined and incubated at room temperature for an additional 20 minutes. The newly formed transfection complexes were subsequently added to the cells and incubated at 37 C with 5% CO 2 for 24 hours, before replacement with fresh serum-free medium before experimentation. As a control, cells were

HYAL2 Governs RhoA Driven Cell Responses
The American Journal of Pathologyajp.amjpathol.org stimulation. Fibroblasts were seeded into 35-mm cell culture plates and grown to confluent monolayers. After 48 hours of growth arrest, cells were incubated with serumfree medium alone (control) or serum-free medium containing 10 ng/mL TGF-b1 for 72 hours. A: Cells were harvested, and RNA was isolated and purified. Quantitative RT-PCR was used to determine mRNA expression of HYAL2. B: Total protein was isolated and immunoblotted for total HYAL2 protein production. Densitometric analysis is shown alongside. Blots are representative of three separate experiments. C: Fibroblasts were seeded into eight-well glass chamber slides and grown to 60% to 70% confluence prior to growth arrest. Cells were treated with serum-free medium alone or serum-free medium containing 10 ng/mL TGF-b1. Cells were washed with phosphate-buffered saline before fixation, permeabilization, and staining to visualize HYAL2 and F-actin expression/localization. Images were captured using confocal laser scanning microscopy and are representative of three individual experiments. Total HYAL2 expression is expressed as a percentage of the total HYAL2 expression by cellular localization. Arrowheads indicate areas of colocalization (yellow). HYAL2 localization quantification: F-actinestained regions (green channels) were isolated by threshold settings using ImageJ software version 1.37c to identify cytoskeleton regions of interest (ROIs). A second ROI tracing the cell boundary was selected by identifying cell membrane localization. Intensity of HYAL2 staining (red channel) in each region was expressed as a percentage fraction of the total HYAL2 staining. The remaining HYAL2 staining was identified as cytoplasmic localization. Five cells per microscopic field were assessed, and a total of three microscopic fields were analyzed per treatment condition. D: Fibroblasts were treated as in B and total HYAL2 protein was co-immunoprecipitated and subjected to immunoblot for b-actin, g-actin, and asmooth muscle actin (SMA). Blots are representative of three separate experiments. E: Cells were treated as in C but with additional treatment of cytochalasin B, followed by immunostaining for HYAL2 and F-actin. Images are representative of three individual experiments. Data are expressed as means AE SEM of three independent experiments. **P < 0.01. Scale bars Z 10 mm (C and E). Original magnification: Â600 (C); Â400 (E). con., control; GADPH, glyceraldehyde phosphate dehydrogenase. ajp.amjpathol.org -The American Journal of Pathology transfected with negative-control siRNA (scramble; a sequence that bears no homology to the human genome) (Thermo Fisher Scientific).

Overexpression Vector Generation and Transfection
The HYAL1 or HYAL2 open reading frame was inserted into the vector pCR3.1 using a standard ligation reaction with T4 DNA ligase (New England BioLabs, Hitchin, UK). Amplification of the cloned vector was performed via bacterial transformation into one-shot competent Escherichia coli (New England BioLabs) and grown overnight on ampicillin containing agar. Single colonies were extracted, cloned, and DNA-purified according to the Miniprep Kit protocol (Sigma-Aldrich). Cloned pCR3.1-HYAL1 or pCR3.1-HYAL2 vector uptake of the insert was confirmed with DNA sequencing. Transfection was done using Lipofectamine LTX transfection reagent (InvitroGen/Thermo Fisher Scientific) in accordance with the manufacturer's protocol. Briefly, cells were grown to 50% to 60% confluence in antibiotic-free medium in either 35-mm dishes or eight-well Permanox chamber slides. Transfection reagent (2% v/v) was diluted in Opti-MEM reduced growth medium (Gibco/Thermo Fisher Scientific). pCR3.1 overexpression vectors were diluted in Opti-MEM reduced growth medium containing PLUS Reagent (1% v/v), to achieve a final transfection concentration of 1.5 mg/mL. The Figure 3 Hyaluronidase (HYAL)-2 from cell lysates (CL) of myofibroblasts has no enzymatic action in hyaluronan (HA)-degradation assays. Fibroblasts were grown to 70% confluence and were transiently transfected with pCR3.1 containing HYAL1 or HYAL2 coding regions. Control transient transfections were performed with the corresponding empty vector. Cells were growth-arrested before incubation in serum-free medium containing 10 ng/mL transforming growth factor beta 1 (TGFB1; TGF-b1). A: Total cellular RNA was extracted, purified, and analyzed by RT-PCR, and the products were separated on 3% agarose gels. Controls were negative (Àve) RT and negative (Àve) PCR, where no polymerase enzyme was added to the reaction mixture. B: HA substrate gel zymography of HYAL activity. The medium was collected, and the cells were extracted in lysis buffer. The HYAL activity was isolated from the conditioned medium (CM) and CL by DEAE-Sephacel ion exchange and lyophilized. Positive controls consisted of diluted human serum. Polyacrylamide gel electrophoresis was performed without SDS in a 7.5% polyacrylamide gel (þHA) and visualized by Alcian Blue and Coomassie Brilliant Blue stain. C: HYAL activity (circles) of CM and CL was further analyzed by incubating with highemolecular-weight [ 3 H]-HA. HA size distribution was examined and compared with the original [ 3 H]-HA preparation, incubated in the absence of the conditioned medium or CL concentrate (squares). All images are representative of four individual experiments. GADPH, glyceraldehyde phosphate dehydrogenase.

HYAL2 Governs RhoA Driven Cell Responses
The American Journal of Pathologyajp.amjpathol.org transfection reagent and plasmid mixtures were then combined and incubated at room temperature for an additional 20 minutes. The newly formed transfection complexes were subsequently added to the cells and incubated at 37 C with 5% CO 2 for 24 hours, before replacement with fresh serum-free medium before experimentation. As a control, cells were transfected with empty pCR3.1 plasmid (mock transfection plasmid containing no open reading frame cDNA). Negative RT experiments were performed alongside HYAL1/HYAL2 mRNA PCR to ensure that the vectors were not conveying falsepositive overexpression. Expression was confirmed by visualization on a 2% agarose gel.

HA Substrate Gel Zymography
Confluent monolayers of fibroblasts were cultured in T25 culture flasks, growth-arrested, and incubated in serum-free medium containing 10 ng/mL TGF-b1 for up to 72 hours. The conditioned medium (CM) was removed, and the cells were extracted in lysis buffer. The CM and the cell lysate (CL) were passed over DEAE-Sephacel ion-exchange columns in 50 mmol/L Tris-HCl, pH 7.8. The flow-through containing the HYAL enzymes was collected, dialyzed against H 2 O, and lyophilized. The samples were reconstituted in 100 mL of H 2 O and mixed with an equal volume Figure 4 Inhibition of hyaluronidase (HYAL)-2 expression delays but does not prevent myofibroblast phenotype acquisition or transforming growth factor beta 1 (TGFB1; TGF-b1)eassociated gene expression. Fibroblasts were grown to 50% confluence prior to transfection with negative control (scrambled) siRNA or with siRNA targeting HYAL2 expression (siHYAL2). Cells were growth-arrested for 48 hours before treatment with serum-free medium alone or serum-free medium containing 10 ng/mL TGF-b1 for 0, 24, 48, or 72 hours. A: Western blot was used to assess HYAL2 protein knockdown. B: Quantitative RT-PCR (RT-qPCR) was used to assess HYAL2 mRNA knockdown. C: Cells were then washed with phosphate-buffered saline, fixed, and permeabilized before staining to visualize a-smooth muscle actin (SMA). Images are representative of three individual experiments. DeF: RT-qPCR was also used to assess TGF-b1eregulated expression of ACTA2 (a-SMA) (D), extra domain-Aefibronectin (EDA-FN) (E), and TGFB1 (F) mRNA after incubation with either HYAL2 or scrambled siRNA. Data are expressed as means AE SEM of three independent experiments. *P < 0.05, **P < 0.01, and ***P < 0.001. Scale bars Z 10 mm.
of zymography loading buffer. Samples were loaded onto SDS-free 7.5% polyacrylamide gels containing HA at a final concentration of 0.17 mg/mL. Positive controls consisted of 1 mL of human serum diluted in 10 mL of H 2 O, mixed with an equal volume of zymography loading buffer. Electrophoresis was performed in running buffer at 75 V for 3 hours on ice to prevent denaturation of the HYAL enzymes.
The gel was incubated in 0.1 mol/L sodium formate buffer, pH 3.7 at 37 C for 72 hours. The gel was stained with 0.5% (wt/v) Alcian Blue and then with 0.1% (w/v) Coomassie Brilliant Blue in 50% (v/v) methanol and 20% (v/v) acetic acid. The gel was destained in 5% (v/v) methanol and 10% (v/v) acetic acid. The gel was visualized and imaged using a light-box.

HYAL2 Governs RhoA Driven Cell Responses
The American Journal of Pathologyajp.amjpathol.org The samples were mixed with an equal volume of 4 mol/L guanidine buffer and analyzed by size-exclusion chromatography on a Sephacryl S-500 column equilibrated with 4 mol/L guanidine buffer or a Sepharose CL-4B column (Amersham Pharmacia Biotech, Buckinghamshire, UK) equilibrated with 4 mol/L guanidine buffer. HYAL activity was assessed by comparing the [ 3 H]-HA elution profiles, incubated in the presence and absence of CM or CL. HYAL activity was not detectable at neutral pH (data not shown).

Co-Immunoprecipitation
Cells were harvested into 1 mL of ice-cold PBS and pelleted at 1000 g for 5 minutes before resuspension in radioimmunoprecipitation assay lysis buffer. Supernatant was transferred to new Eppendorfs and a known volume of 25 mg of protein in radioimmunoprecipitation assay buffer was immunoprecipitated using anti-HYAL2 antibody-linked anti-rabbit Dynabeads (InvitroGen/ Thermo Fisher Scientific). Beads were preincubated with 0/1% w/v BSA. Immunoprecipitation was completed with an overnight rotating incubation at 4 C. After three washes with Nonidet P-40 buffer, the beads were resuspended in PBS and transferred to clean microcentrifuge tubes. The bead-antibody-protein complex was boiled with reducing buffer for 5 minutes before the supernatant was transferred into gel lanes for SDS-PAGE. Alternatively, eluted protein was assessed by mass spectrometry (MS).

Mass Spectrometry
Co-immunoprecipitation elutes were loading into and run on a 1.5-mm 7.5% polyacrylamide gel. Gel plugs were manually excised and peptides recovered after trypsin (6.25 ng/mL in 25 mmol/L NH 4 HCO 3 , 37 C, 3 hours; sequencing gradeemodified trypsin from Promega UK Ltd., Southampton, UK) digestion using a modified version of the method of Shevchenko et al. 44 The dried peptides were resuspended in 50% (v/v) acetonitrile in 0.1% (v/v) trifluoroacetic acid (5 mL

Collagen Gel Contraction Assays
Type I collagen was extracted from rat-tail tendon as previously described. 47 Approximately 2.5 Â 10 5 fibroblasts/mL were mixed into collagen latticeeforming solutions (2.5 mL 20% v/v fetal calf serumeDulbecco's modified Eagle's lowglucose medium, 500 mL of 0.1 mol/L NaOH, and 2 mg/mL type I collagen; total volume of 5 mL). Fibroblast-populated collagen lattices (FPCLs) were maintained at 37 C, in a 5% CO 2 atmosphere for 1 hour, for collagen polymerization to occur. FPCLs were gently detached from the plate edges and resuspended in serum-free medium containing appropriate

HYAL2 Governs RhoA Driven Cell Responses
The American Journal of Pathologyajp.amjpathol.org treatments. FPCLs were measured at 0, 3, 6, 12, and 24 hours after initial lattice fabrication. The mean FPCL contraction values were obtained from analysis by ImageJ software version 1.37c (and are expressed as the percent reduction in gel diameter compared to the gel diameter at 0 days).

Scratch-Wound Migration Assays
Scratching quiescent fibroblasts cell monolayers with sterile 200-mL pipette tips generated linear denuded areas. The cells were gently washed with PBS to remove Fibroblasts were grown to 50% confluence prior to transfection with negative control (scramble) siRNA (À) or with siRNA targeting HYAL2 expression (siHYAL2) (þ). After successful transfection and growth arrest, cells were incubated with 10 ng/mL transforming growth factor (TGF)-b1 for the annotated times. At each time point, cells were harvested and protein from total cell lysate was isolated. Phosphorylation of RhoA and MLCK were determined by Western blot. Total RhoA and MLCK were used as gel loading controls. B: The RhoA inhibitor Rhosin was used as an inhibitor of RhoA activation. Fibroblasts were pretreated with 10 mmol/L Rhosin for 2 hours, prior to and during incubation with 10 ng/mL TGF-b1. Phosphorylation of RhoA and MLCK were determined by Western blot. C: Fibroblasts were grown to confluence in culture and after 48 hours of growth arrest were incubated with either serum-free medium containing 10 ng/mL TGF-b1 or serum-free medium alone (control). Anti-HYAL2 co-immunoprecipitation followed by Western blot for MLCK, RhoA, and calcium/calmodulin-dependent protein kinase type II (CaMKII) was then performed. Western blots for HYAL2 were used as loading controls. Positive (þve) controls were total cell lysate and negative (Àve) controls were co-immunoprecipitation (IP) performed using rabbit Immunoglobulin G (IgG) in place of anti-HYAL2 antibody. All blots are representative of three independent experiments. Densitometries are expressed as the means AE SEM of three independent experiments. *P < 0.05, **P < 0.01, and ***P < 0.001., con. control; min., minutes; p, phosphorylated. ajp.amjpathol.org -The American Journal of Pathology detached cells and then replenished with fresh serum-free medium containing appropriate treatments. The wound size was photographed at 0, 3, 6, 12, and 24 hours or until closure, using an Axiovert 100 mol/L inverted microscope (Carl Zeiss, Oberkochen, Germany) fitted with a digital camera (Orca-1394; Hamamatsu Photonics K.K., Hamamatsu, Japan). Measurements were obtained using ImageJ software version 1.37c. Data are expressed as the percent reduction in wound area, compared to wound area at 0 hours.

Statistical Analysis
The two-tailed, unpaired t-test was used to assess statistical differences between the two experimental groups. For experiments with multiple experimental groups, one-way analysis of variance was used to identify statistical differences across groups, followed by post-test Bartlett and multiple comparisons. For experiments with multiple experimental conditions, two-way analysis of variance was used, followed by post-test Tukey multiple comparisons. Graphical data are expressed as means AE SEM. All data were analyzed using GraphPad Prism software version 6 (GraphPad Software, San Diego, CA). P < 0.05 was considered statistically significant.

Increased HYAL2 Expression and Colocalization with a-SMAePositive Myofibroblasts in the Renal Interstitium after Experimental Renal Fibrosis in Vivo
To investigate the in vivo relevance of HYAL2 in progressive fibrosis, we characterized HYAL2 expression, localization, and preponderance in relation to the expression of the myofibroblast marker a-Sma in an experimental model of renal fibrosis (Figure 1). Our established model of bilateral renal ischemia was used to promote renal fibrosis in rats as described in the methods, and comparisons were made to sham controls. The results demonstrated that in normal/sham kidneys ( Figure 1A), HYAL2 staining was present only in arterial blood vessels, where it colocalized with a-Sma staining from vascular smooth muscle cells.
There was little/no HYAL2 staining in the glomeruli or in the tubules. There was some nonspecific HYAL2 staining of red blood corpuscles in the glomeruli and interstitium consistent red cell autofluorescence, which is a common phenomenon in the kidney. The renal interstitium is the space between the renal tubules and glomeruli where the fibroblasts/stromal cells reside. Resident fibroblasts were seen in the renal interstitium; however, these demonstrated

HYAL2 Governs RhoA Driven Cell Responses
The American Journal of Pathologyajp.amjpathol.org no a-Sma expression, indicating that these cells had not differentiated to a profibrotic myofibroblast phenotype. The resident fibroblasts demonstrated little/no HYAL2 expression.
Kidneys harvested from rats that had undergone ischemia reperfusion injuryeinduced renal fibrosis ( Figure 1B) also demonstrated positive HYAL2 staining in arterial blood vessels, and this similarly colocalized with a-Sma expression from vascular smooth muscle cells. However, in these kidney sections there was also clear evidence of increased expression of a-Smaepositive myofibroblasts in the renal interstitium. The areas of interstitial a-Smaepositive staining also demonstrated increased HYAL2 expression, which colocalized with a-Smaepositive areas. The top panel of images in Figure 1B (from rodents that underwent ischemia reperfusion injury) depicts an area of fibrosis with relatively preserved renal architecture with undamaged renal tubules. In these images there is a-Smaepositive myofibroblast staining with associated HYAL2 staining and colocalization of these two proteins in the renal interstitium. The lower panel of images in Figure 1B depicts an area of gross fibrosis with disordered renal tubular architecture and increased a-Smaepositive myofibroblast staining in the renal interstitium. These areas of enhanced fibrosis and damage also demonstrated increased levels of HYAL2 staining and colocalization with a-Smaepositive myofibroblasts. These data indicate that during progressive fibrosis, there is increased differentiation to and expansion of the a-SMAepositive myofibroblast cell population in the renal stroma, and that HYAL2 expression is increased in this cell population and colocalizes with a-SMAepositive cytoskeleton.

Increased HYAL2 Expression in Myofibroblasts Associated with the Cell Cytoskeleton
Exposure of human fibroblasts to 10 ng/mL TGF-b1 for 72 hours was previously demonstrated to induce terminal myofibroblast differentiation. 14,43 Fibroblasts were stimulated with this established dose and duration of TGF-b1 to induce stable myofibroblast differentiation and expression of HYAL2 mRNA assessed by qPCR. Myofibroblasts had significantly increased expression of HYAL2 mRNA compared to undifferentiated fibroblasts (Figure 2A), and this finding was reciprocated on detection of total HYAL2 protein present in CLs ( Figure 2B). The cellular localization of HYAL2 was next determined. Immunofluorescence staining for HYAL2 and filamentous actin demonstrated that in undifferentiated fibroblasts, HYAL2 was predominantly localized at the cell surface, while in differentiated knockdown. Fibroblasts were grown to 50% confluence prior to transfection with negative control (scramble) siRNA or with siRNA targeting HYAL2 expression (siHYAL2). Cells were subsequently growth-arrested before treatment with serum-free medium containing 10 ng/mL transforming growth factor beta 1 [TGFB1, also known as (TGF)-b1], for 0, 24, 48, or 72 hours. Quantitative RT-PCR was used to assess the effects of HYAL2 knockdown on mRNA expression of cellular communication network factor 2 (CTGF) (A), fibronectin 1 (FN1) (B), matrix metallopeptidase 2 (C), collagen type I alpha 1 chain (COL1A1) (D), and collagen type I alpha 2 chain (COL1A2) (E). Data are expressed as means AE SEM of three independent experiments. *P < 0.05, **P < 0.01. ajp.amjpathol.org -The American Journal of Pathology myofibroblasts, HYAL2 was localized along and around the actin cytoskeleton ( Figure 2C). To confirm the nature of HYAL2 association with the actin cytoskeleton in myofibroblasts in culture, the cells were assessed for coimmunoprecipitation of cytoskeletal components including b-actin, g-actin, and a-SMA in undifferentiated and TGF-b1edifferentiated myofibroblasts. The immunoblots indicated that after TGF-b1edriven myofibroblast differentiation there was increased association between HYAL2 and a-SMA ( Figure 2D). To further confirm this finding, cells were treated with cytochalasin B to prevent actin filament polymerization prior to immunofluorescence for HYAL2. Disruption of the actin cytoskeleton by preventing filamentous actin polymerization also led to the disruption of cytoplasmic HYAL2, confirming that HYAL2 associates with the a-SMA cytoskeleton in myofibroblasts ( Figure 2E).

HA Catabolic Activity of Cell-Associated HYAL2 in Myofibroblasts
To determine the catabolic effects of HYAL2 in myofibroblasts, these cells were transiently transfected with plasmid vector (pCR3.1) containing the HYAL2 coding region. To compare catabolic activity of HYAL2 with the catabolic activity of the other widely expressed HYAL enzyme in vertebrates, HYAL1, fibroblasts were also transfected with plasmid vector (pCR3.1) containing the HYAL1 coding region. Semi-quantitative RT-PCR was used to examine the mRNA expression of HYAL1 and HYAL2 in fibroblasts transiently transfected with pCR3.1-HYAL1 and pCR3.1-HYAL2. Control transfection was performed with the empty vector pCR3.1. Endogenous HYAL1 and HYAL2 mRNA expression was detected in control transfections. Transfection with pCR3.1-HYAL1 resulted in increased HYAL1 mRNA expression compared to transfection with the empty vector. Transfection with pCR3.1-HYAL2 resulted in increased HYAL2 mRNA expression compared to the empty vector ( Figure 3A). HA substrate gel zymography of HYAL activity of fibroblasts transiently transfected with pCR3.1-HYAL1 and pCR3.1-HYAL2 was undertaken. Control transfections were undertaken with the empty vector pCR3.1. The HYAL activity was isolated from the CM and CL by DEAE-Sephacel ion-exchange and lyophilized. The HYAL activity was analyzed by HA-substrate gel zymography at pH 3.7. HYAL activity was represented as a clear band in the polyacrylamide gel, due to the exclusion of the Alcian Blue and Coomassie Brilliant Blue stains. Endogenous HYAL activity was not detected in the CM or CL of control transfections. Transfection with pCR3.1-HYAL1 resulted in significant HYAL activity detected in both the CM and CL. However, transfection with pCR3.1-HYAL2 did not result in any detectable HYAL activity in either CM or CL ( Figure 3B). To confirm these results, cells transiently transfected with either pCR3.1 alone, pCR3.1-HYAL1, or pCR3.1-HYAL2 were isolated from the CM and CL by DEAE-Sephacel ion-exchange chromatography and lyophilized. The CM and CL were incubated with highemolecular-weight [ 3 H]-HA at pH 3.7 for 72 hours. The samples were then subjected to size-exclusion chromatography and HYAL activity assessed by comparing the elution profiles of the [ 3 H]-HA incubated in the presence or absence of CM or CL. Endogenous HYAL activity was detected in the CM of the control transfections with pCR3.1 alone, partially degrading the HA to loweremolecular-weight forms. Transfection with pCR3.1-HYAL1 increased the HYAL activity detected in the CM compared to the control transfection. The CL contained relatively less HYAL activity, only partially degrading the HA. Transfection with pCR3.1-HYAL2 increased the HYAL activity in the CM compared to control transfections with pCR3.1 alone. HYAL activity was not detected in the CL of cells Figure 9 Ras homolog family member A (RhoA) inhibition does not attenuate hyaluronidase 2 (HYAL2) mRNA or hyaluronidase (HYAL)-2 protein expression. Fibroblasts were grown to confluent monolayers on 35-mm tissue culture plates. They were then growth-arrested and pretreated with either dimethyl sulfoxide (DMSO) alone (as a control) or with 10 mmol/L Rhosin for 2 hours, prior to and during incubation with 10 ng/mL transforming growth factor beta 1 [TGFB1, also known as (TGF)-b1] for 72 hours. A: Quantitative RT-PCR was used to assess the effect of RhoA inhibition on HYAL2 mRNA expression. B: Western blot was used to assess the effect of RhoA inhibition on HYAL2 protein expression. GAPDH, glyceraldehyde phosphate dehydrogenase.

HYAL2 Involvement in Mediating TGF-b1eDriven Myofibroblast Differentiation
Myofibroblasts demonstrate cytoskeletal reorganization compared to undifferentiated fibroblasts. Fibroblasts generally demonstrate a narrow, spindle-shaped appearance with actin fibers forming a complex cortical meshwork at the periphery of cells. In myofibroblasts, the actin fibers coalesce to form thick parallel stress fibers that run from end to end of the cells, and the contractile protein a-SMA is incorporated into these stress fibers. 34,35 In view of the results from Figures 1 and 2 indicating that HYAL2 demonstrated association with and relocalization to associate with the actin cytoskeleton as fibroblasts underwent differentiation to myofibroblasts, a link between HYAL2 and myofibroblast differentiation was investigated. siRNAs targeting HYAL2 (siHYAL2) were used to determine the role of this protein in TGF-b1edriven myofibroblast differentiation. Successful knockdown of HYAL2 mRNA and HYAL2 protein expression, after siRNA treatment compared with scrambled controls, was initially confirmed (Figure 4, A and B). The influence of HYAL2 knockdown on a-SMA protein expression was subsequently assessed. There were no significant effects of siHYAL2 transfection versus control (scrambled siRNA transfection) on a-SMA incorporation into myofibroblast stress fibers as visualized through immunofluorescence ( Figure 4C). The effects of HYAL2 knockdown on ACTA2 (a-SMA) mRNA expression assessed by qPCR similarly demonstrated no effects on terminal myofibroblast Figure 10 Ras homolog family member A (RhoA) regulates transforming growth factor beta 1 (TGFB1, also known as TGF-b1)edependent contractility and migration in myofibroblasts. Fibroblasts were grown to confluent monolayers, growth-arrested for 48 hours, and then treated with 10 mmol/L Rhosin for 2 hours prior to and during TGF-b1 incubation. A: Cells were seeded into collagen gels and left to polymerize before incubation in serum-free medium alone or serum-free medium containing 10 ng/mL TGF-b1. Collagen gels were imaged over the annotated time points and measured for analysis of rate of contraction. Dotted lines indicate the measured gel areas at this time point. B: Cells were growth-arrested and scratched using a 20 mL pipette tip. After a phosphatebuffered saline wash, the medium was replaced with fresh serum-free medium or serum-free medium containing 10 ng/mL TGF-b1. Scratch assays were imaged at the indicated time points, and the area of closure was measured to assess rate of migration. Dotted lines represent the original scratch edges. Data are expressed as means AE SEM of three individual experiments. *P < 0.05, **P < 0.01, and ***P < 0.001 control þ TGF-b1 versus treated þ TGF-b1; y P < 0.05, yy P < 0.01 control versus treated samples; zzz P < 0.001 control versus control þ TGF-b1 samples. DMSO, dimethyl sulfoxide. ajp.amjpathol.org -The American Journal of Pathology differentiation at 72 hours ( Figure 4D). Although there was a delay in increased ACTA2 mRNA expression at 24 and 48 hours, at 72 hours after TGF-b1 stimulation ACTA2 mRNA expression in fully differentiated myofibroblasts was comparable in siHYAL2-treated versus scrambled-siRNAetreated cells. EDA splice variant FN is another characteristic myofibroblast marker necessary for acquisition of this profibrotic phenotype. 48,49 The influence of siHYAL2 transfection on EDA-FN mRNA expression also demonstrated that while HYAL2 knockdown delayed increased EDA-FN mRNA expression, it did not influence EDA-FN mRNA expression in differentiated myofibroblasts ( Figure 4E). HYAL2 knockdown also did not influence autoinduction of TGFB1 mRNA expression in myofibroblasts, indicating that overall HYAL2 did not influence TGF-b1edependent terminal myofibroblast differentiation ( Figure 4F).

HYAL2 Binding to RhoA and Myosin Light-Chain Kinase and Modulation of Phosphorylation of These Proteins
MS was used to identify important HYAL2 cellular interactions in myofibroblasts. The Mascot score was used to justify the accuracy of protein identification, and Table 2 lists the proteins identified in descending order of significance. HYAL2 was identified as associating with actin-related cytoskeletal remodeling proteins (actinrelated protein T1, b-actin, g 1 -actin), contractile proteins (myosin heavy chain 9), prostate, ovary, testis-expressed ankyrin domain family members E/F; vinculin; and LIM and SH3 domain protein 1. Table 3 shows the GO and InterPro results from uploading the MS hits and demonstrates the lists of relevant pathways with these hits. To summarize the proteomics results into a general functional model, the known and predicted proteineprotein Figure 11 Hyaluronidase (HYAL2, also known as HYAL-2) mediates Ras homolog family member A (RhoA)-dependent contractility and migration in myofibroblasts. Fibroblasts were grown to 50% confluence prior to transfection with negative control (scramble) siRNA or with siRNA targeting HYAL2 expression (siHYAL2). A: Cells were seeded into collagen gels and left to polymerize before incubation in serum-free medium alone or serumfree medium containing 10 ng/mL transforming growth factor (TGF)-b1. Collagen gels were imaged over the annotated time points and measured for analysis of rate of contraction. Lines indicate the measured gel areas at this time point. B: Cells were growth-arrested and scratched using a 20-mL pipette tip. After a phosphate-buffered saline wash, the medium was replaced with fresh serum-free medium or serum-free medium containing 10 ng/mL TGF-b1. Scratch assays were imaged at the indicated time points, and the area of closure was measured to assess the rate of migration. Lines represent the original scratch edges. Data are expressed as the means AE SEM of three individual experiments. *P < 0.05, **P < 0.01, and ***P < 0.001 control þ TGF-b1 versus treated þ TGF-b1; yy P < 0.01 control versus treated samples; z P < 0.05, zz P < 0.01, zzz P < 0.001 control versus control þ TGF-b1 samples.

HYAL2 Governs RhoA Driven Cell Responses
The American Journal of Pathologyajp.amjpathol.org interactions were built using the STRING database network and demonstrated that all of the proteins are closely related, providing additional evidence that HYAL2 is involved in pathways consisting of these proteins ( Figure 5A). Of the suggested pathways, actin cytoskeleton regulation was selected because of the observations demonstrating that HYAL2 aligns with the cell cytoskeleton ( Figure 5B). Figure 5B also demonstrates the strength of HYAL2eprotein associations within that pathway. This analysis indicates a putative link between HYAL2, RhoA, and MLCK.
To conclusively determine the association between the HYAL2 and RhoA pathways, fibroblasts stimulated with TGF-b1 were transfected with either siHYAL2 or scrambled siRNA. Subsequent Western blot analysis to assess RhoA and MLCK phosphorylation was undertaken and compared to total RhoA and MLCK, respectively, as a control. The results demonstrated that HYAL2 knockdown significantly attenuated RhoA phosphorylation at all time points after TGF-b1 stimulation. HYAL2 knockdown also attenuated MLCK phosphorylation at 15, 30, 60, and 120 minutes after TGF-b1 stimulation. These results were confirmed by densitometric analysis of Western blot bands indicating statistically significant attenuation of RhoA and MLCK activation after HYAL2 knockdown ( Figure 6A). To determine the signaling of the link between RhoA and MLCK, a chemical inhibitor of RhoA (Rhosin) was used in subsequent experiments. The data indicated that Rhosin effectively attenuated RhoA phosphorylation. Furthermore, RhoA inhibition with Rhosin also abrogated MLCK phosphorylation at 15, 30, 60, and 120 minutes after TGF-b1 stimulation, and this was confirmed on densitometric analysis ( Figure 6B). Co-immunoprecipitation was used to examine any direct interaction between these proteins, and the results indicated that HYAL2 interacts directly with RhoA in both fibroblasts and myofibroblasts. HYAL2 also interacts with MLCK, and this interaction is increased after stimulation with TGF-b1. Previous work has demonstrated an important role for calcium/calmodulin-dependent protein kinase type II signaling in mediating TGF-b1edriven fibroblast to myofibroblast differentiation. 42 This study therefore also tested for direct interactions between HYAL2 and calcium/calmodulin-dependent protein kinase type II and identified no association between these two proteins ( Figure 6C).

Attenuated Expression of RhoA-Dependent Genes after HYAL2 Knockdown
RhoA is a small GTPase protein in the Rho family, which has been implicated in regulating several fibrogenic genes including collagens, FNs, matrix metalloproteinases (MMPs), and connective tissue growth factor [CTGF; also known as cellular communication network factor (CCN2)]. 50e54 To confirm the involvement of RhoA in the regulation of these genes in our experimental cell systems, fibroblasts were incubated with 10 ng/mL TGF-b1 for up to 72 hours to induce myofibroblast differentiation and treated with either a chemical RhoA inhibitor (Rhosin) or incubated with dimethyl sulfoxide alone. The effects of RhoA inhibition on mRNA expression of CCN2, FN1, MMP2, and COL1A1/COL1A2 (collagen type 1 a chains) were subsequently assessed using qPCR. The results demonstrated that RhoA inhibition attenuated CTGF, FN1, MMP2, COL1A1, and COL1A2 mRNA expression, indicating that RhoA regulates expression of these genes in myofibroblasts (Figure 7). The effects of HYAL2 knockdown on these RhoA-regulated genes was then investigated ( Figure 8). CTGF mRNA expression was significantly attenuated upon HYAL2 knockdown at all timepoints after TGF-b1 stimulation ( Figure 8A). Expression levels of FN1, MMP2, COL1A1, and COL1A2 all demonstrated significant attenuation after HYAL2 knockdown compared to scrambled controls in cells that had differentiated to myofibroblasts (Figure 8, BeE). Furthermore, there was no effect on HYAL2 mRNA ( Figure 9A) or HYAL2 protein ( Figure 9B) expression after Rhosin treatment, suggesting that RhoA inhibition did not directly influence HYAL2 expression.

HYAL2 Regulation of RhoA-Mediated Contractility and Migration in Myofibroblasts
The activity of Rho GTPases in regulating the polymerization and organization of actin and myosin filaments has been previously established. 52 The role of RhoA in regulating cytoskeletal-related myofibroblast functions was therefore assessed in our experimental cell systems. The effect of RhoA on TGF-b1edriven activated fibroblast contractility was initially assessed. The results demonstrated that TGF-b1 enhanced collagen type I gel contraction, whereas treatment with the RhoA inhibitor (Rhosin) abrogated TGF-b1edriven activated fibroblast contractility ( Figure 10A). In addition, RhoA inhibition also markedly attenuated TGF-b1edriven migration in these cells ( Figure 10B). To determine the role of HYAL2 in regulating TGF-b1edriven activated fibroblast contractility and migration, fibroblasts were transfected with either scrambled siRNA or siHYAL2 and subsequently incubated with 10 ng/mL TGF-b1. Similar to the effects seen with RhoA inhibition, knockdown of HYAL2 reduced TGF-b1-driven fibroblast contractility and significantly suppressed migration, over the course of 24 hours ( Figure 11). Thus, RhoA and HYAL2 are both involved in mediating pathways associated with TGF-b1edriven activated fibroblast contractility and migration.

Discussion
HYAL2 is a 54-kDa protein that has high-level expression in most tissues, including skin, liver, kidneys, skeleton, heart, and lungs, suggesting that it has an important biological role. ajp.amjpathol.org -The American Journal of Pathology However, while the primary function of HYAL2 is considered to be enzymatic, it is only a weak HA-degrading enzyme. Hence, although recent studies have indicated that mutations in HYAL2 lead to significant cardiac, skeletal, hematopoietic, and other abnormalities, 15,16,18,20 the cellular mechanisms through which HYAL2 achieves its effects and regulates biology are poorly understood.
The findings from this study, in conjunction with other recently published studies, indicate that HYAL2 has important and previously undescribed nonenzymatic effects that influence distinct and opposing fibroblastic cell functions. Previous studies demonstrated the role of nuclear HYAL2 as a key regulator of CD44 alternative splicing, and in particular, highlighted its function in promoting the expression of an antifibrotic CD44 splice variant termed CD44v7/8. 27,43 Cell-surface expression of this CD44 splice variant in fibroblasts resulted in prevention and reversal of TGF-b1edriven myofibroblast differentiation, thereby limiting profibrotic tissue damage. In contrast, the results presented here demonstrate that cytoplasmic HYAL2 promotes profibrotic processes. This study shows that TGF-b1 promotes HYAL2 relocalization to the cytoplasm, where it binds to the actin cytoskeleton. HYAL2 in this context modulates RhoA signaling, promoting downstream profibrotic myofibroblast functions. Specifically, HYAL2 enhances RhoA-dependent cell migration and, to a lesser extent, contractility. Many studies demonstrate RhoA to be promigratory, while some studies report RhoA to be antimigratory. 52,55,56 While this dichotomy is not fully understood, the findings from this study indicate that HYAL2eRhoA interactions are crucial in regulating RhoAdependent migration and may explain its divergent functions. HYAL2 also promotes increased collagen, FN, CTGF (CCN2) and MMP2 expression. In these studies, despite having low HA catabolic activity, HYAL2 influences wideranging cellular and matrix effects through its influence on myofibroblast phenotype and function. Given the central role of myofibroblasts in regulating homeostasis and driving pathology, the broad expression of HYAL2 in tissues and the widespread impact of its dysregulation on homeostasis and disease are now more understandable. Whether HYAL2 has similar nonenzymatic actions in nonstromal cell populations remains to be investigated.
HYAL2 requires a pH optimum of 4.0 for its HAbreakdown activity. 22 Little is known regarding the factors that may govern the switch from enzymatic to nonenzymatic HYAL2 functions. At the cell surface, glycosylphosphatidylinositol-linked HYAL2 has been shown to interact with the Na þ /H þ exchanger 1. 57 Na þ /H þ exchanger 1 has been proposed to create acidic microenvironments at the cell surface accounting for any membranebound HYAL2 catalytic activity. Furthermore, cell-surface HYAL2 catalytic activity has previously been found to be dependent on the expression of cell-surface CD44. 58 Nuclear HYAL2 has been found to be associated with a range of RNA-and DNA-processing proteins and influences the incorporation of serine-and arginine-rich splice factors and snRNAs into the spliceosome. 27 HYAL2 has also been identified intracellularly within lysosomes and endosomes and now binding to the cytoskeleton. 43,59 At each of these cell locations, HYAL2 was identified as having distinct cellular functions, and it is possible to speculate that the distinct actions of HYAL2 are influenced by its cellular localization. It may be that the nonenzymatic functions of HYAL2 predominate when the protein is localized in a nonacidic microenvironment and detached from CD44 in support of findings from previous studies. 58 The mechanisms that influence HYAL2 cellular localization are as yet unknown, and the proteineprotein interactions that may influence HYAL2 trafficking between the cell surface, cytoplasm, and nucleus will be the focus of further study.
HYAL2 is identified as having a catalytic domain, an HA-binding domain, a glycosylphosphatidylinositol-anchor domain, and an epidermal growth factorelike domain. However, the other putative functional domains of HYAL2 that may regulate the recently identified nonenzymatic functions remain undescribed. It is known that HYAL2 is involved in a number of proteineprotein interactions, including HYAL2eCD44 coupling, HYAL2eWW domainecontaining oxidoreductase 1 and HYAL2ecell migrationeinducing and HA-binding protein interactions. 23,24,60 In these studies, HYAL2 interactions with these particular proteins influence the effects of HYAL2, subsequently influencing distinct cellular functions. As an example, HYAL2 interaction with cell migrationeinducing and HA-binding protein (formerly known as KIAA1199) is thought to enhance the catalytic activity of HYAL2. 60 Hence, it is also possible that as-yeteunidentified proteins may influence a conformational change in HYAL2 structure that is crucial for its various nonenzymatic actions. In the field of fibrosis research, the generation of catalytically inactive mutants of HYAL2, and an understanding of both HYAL2 structural biology and the distinct peptide domains and HYAL2eprotein interactions that could promote antifibrotic versus profibrotic cellular functions, offer an important area of research.
In this study we demonstrated that while cytoskeletallinked HYAL2 did not mediate TGF-b1edriven myofibroblast differentiation, it regulated some very important myofibroblast functions that ultimately influence fibrogenesis. Migration and contractility are crucial myofibroblast functions facilitating scarring. 61 Increased generation of collagens and FNs are hallmarks of fibrotic matrix production by myofibroblasts. In addition, numerous studies have established a link between increased expression of CTGF (CCN2) and MMP2 in fibrotic disease. 53,62 All of these functions are downstream effects of RhoA signaling, as indicated in the experiments using the RhoA inhibitor Rhosin. 51,54 Given the observed direct interaction between HYAL2 and RhoA demonstrated in this study, it appears that cytoskeletal-linked HYAL2 is involved in the orchestration of RhoA signaling and is likely to influence all HYAL2 Governs RhoA Driven Cell Responses The American Journal of Pathologyajp.amjpathol.org downstream RhoA effects. Furthermore, as there seems to be a basal interaction between HYAL2 and RhoA, even in resting/unstimulated fibroblasts, it is speculated that after TGF-b1 stimulation, both HYAL2 and RhoA move together from the cell membrane to the cytoskeleton, where they can both associate with and activate MLCK. In conclusion, HYAL2 in myofibroblasts associates with the actin cytoskeleton and interacts with cytoskeletal-associated RhoA, and this interaction is essential for mediating downstream profibrotic RhoA-signaling effects.