Peroxidase Enzymes Regulate Collagen Extracellular Matrix Biosynthesis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

The American Journal of Pathology, Vol. -, No. -, - 2015

ajp.amjpathol.org

Peroxidase Enzymes Regulate Collagen Extracellular Matrix Biosynthesis Q52

Mark O. DeNichilo,* Vasilios Panagopoulos,* Timothy E. Rayner,y Romana A. Borowicz,z John E. Greenwood,x and Andreas Evdokiou*{ From the Breast Cancer Research Unit,* Discipline of Surgery, the Mesenchymal Stem Cell Laboratory,z School of Medical Sciences, Faculty of Health Sciences, and the Center for Personalized Cancer Medicine,{ The University of Adelaide, Adelaide, South Australia; the Paramedic Unit,y Flinders Clinical Effectiveness, School of Medicine, Flinders University, Bedford Park, South Australia; and the Adult Burn Centre,x Royal Adelaide Hospital, Adelaide, South Australia, Australia Accepted for publication January 21, 2015.

Q3

Q4

Address correspondence to Mark O. DeNichilo or Andreas Evdokiou, Basil Hetzel Research Institute, TQEH, 27a Woodville Rd, Woodville, SA 5011, Australia. E-mail: mark. [email protected] or andreas.evdokiou@adelaide. edu.au.

Myeloperoxidase and eosinophil peroxidase are heme-containing enzymes often physically associated with fibrotic tissue and cancer in various organs, without any direct involvement in promoting fibroblast recruitment and extracellular matrix (ECM) biosynthesis at these sites. We report herein novel findings that show peroxidase enzymes possess a well-conserved profibrogenic capacity to stimulate the migration of fibroblastic cells and promote their ability to secrete collagenous proteins to generate a functional ECM both in vitro and in vivo. Mechanistic studies conducted using cultured fibroblasts show that these cells are capable of rapidly binding and internalizing both myeloperoxidase and eosinophil peroxidase. Peroxidase enzymes stimulate collagen biosynthesis at a post-translational level in a prolyl 4-hydroxylaseedependent manner that does not require ascorbic acid. This response was blocked by the irreversible myeloperoxidase inhibitor 4-amino-benzoic acid hydrazide, indicating peroxidase catalytic activity is essential for collagen biosynthesis. These results suggest that peroxidase enzymes, such as myeloperoxidase and eosinophil peroxidase, may play a fundamental role in regulating the recruitment of fibroblast and the biosynthesis of collagen ECM at sites of normal tissue repair and fibrosis, with enormous implications for many disease states where infiltrating inflammatory cells deposit peroxidases. (Am J Pathol 2015, -: 1e13; http://dx.doi.org/10.1016/j.ajpath.2015.01.013)

Ascorbic acid (AA) has long been recognized as an essential cofactor required for collagen homeostasis, with long-term ascorbate dietary deficiency leading to the inadequate connective tissue renewal characteristic of scurvy.1 AA increases the secretion of collagen proteins by promoting efficient hydroxylation of peptidyl proline, leading to the assembly of a stabilized procollagen triple helical structure.2 Without hydroxylation, unfolded procollagen is thermally unstable and is retained within the cell and degraded.3 Prolyl 4hydroxylase (P4H) is the iron-containing enzyme responsible for catalyzing hydroxylation of procollagen within the cell. AA is thought to act as an electron donor that maintains the iron within the P4H molecule in the Fe2þ state, thus enabling the enzyme to repeatedly hydroxylate proline residues.4 Unlike humans, mice are capable of synthesizing AA in the liver. By genetically deleting key genes responsible for the production of AA in mice, two recent studies concluded

that collagen biosynthesis during both tumor development and renal fibrosis proceeds in the absence of AA, suggesting that alternative, yet unidentified, electron donor sources likely exist. These sources substitute for the requirement of AA in situations where a rapid escalation in collagen biosynthesis is a defining feature.5,6 Myeloperoxidase (MPO) and eosinophil peroxidase (EPO) are heme-containing enzymes whose functional involvement in human health has mainly been studied in the context of providing a mechanism for oxidative defense against invading bacteria and other pathogenic microorganisms.7 Supported in part by The Hospital Research Foundation and Australian Q2 Breast Cancer Research, National Health and Medical Research Council (NHMRC) Career Development Fellowship 627015 (A.E.), and NHMRC Project grant 1050694 (A.E.). Disclosures: None declared.

Copyright ª 2015 American Society for Investigative Pathology. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajpath.2015.01.013

FLA 5.2.0 DTD
AJPA1979_proof
7 March 2015
5:33 pm
EO: AJP14_0571

63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124

125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186

DeNichilo et al Released by infiltrating neutrophils and eosinophils, respectively, these endogenous peroxidases have often been associated with fibrotic tissue in various organs without any direct involvement in fibroblast recruitment or extracellular matrix (ECM) biosynthesis attributed to their presence. For example, elevated accumulation of MPO is associated with augmented fibrosis seen in patients with atrial fibrillation,8 whereas EPO has been implicated in the pathogenesis of human endometriosis9 and experimental ulcerative colitis,10 where focally intense deposition of EPO localizes within fibrotic connective tissue. To date, the capacity of peroxidase proteins to induce a response involving fibroblastic recruitment and the synthesis and deposition of collagen-rich ECM has not been contemplated. Herein, we report studies that characterize, for the first time, the ability of mammalian MPO and EPO as well as plant-derived peroxidase proteins, such as horseradish peroxidase (HRP) and soybean peroxidase (SBP), to directly stimulate the migration and secretion of collagen I and collagen VI by fibroblasts in vitro and, ultimately, promote a fibrogenic response in vivo. We provide mechanistic data that show peroxidase enzymes regulate collagen I biosynthesis at a post-translational level that is P4H dependent, but more important, does not require AA. Our studies demonstrate that the catalytic activity of MPO and EPO is essential to support collagen I biosynthesis by promoting P4H-dependent hydroxylation and subsequent release of collagen from cells. Our novel findings that peroxidase enzymes, such as MPO and EPO, can substitute for AA to promote collagen biosynthesis strongly suggest that these enzymes play an important role not only in normal tissue repair, but also in the progression of disease states where infiltrating inflammatory cells are known to deposit peroxidases.

Materials and Methods Peroxidases Native human eosinophil peroxidase (EPO) was obtained from Cell Sciences (Canton, MA) and Lee Biosolutions Inc. (St. Louis, MO). HRP was purchased from Sigma-Aldrich (Sydney, NSW, Australia). Recombinant human MPO was purchased from R&D Systems (Minneapolis, MN). SBP was obtained from Bio-Research Products, Inc. (North Liberty, IA).

Fibroblast Cell Culture

Q5

Primary human fibroblasts were grown in Dulbecco’s modified Eagle’s medium (DMEM; high glucose with no AA), supplemented with 2 mmol/L glutamine, 100 IU/mL penicillin, 100 mg/mL streptomycin, 20 mmol/L HEPES, and 10% fetal bovine serum (FBS; Invitrogen) in a 5% CO2-containing humidified atmosphere. Human foreskin fibroblasts (HFFs) were isolated and provided by Dr. Pritinder Kaur (Peter Mac Callum Institute, Melbourne, VIC, Australia) and used

2

between cell passages 8 and 10. Adult dermal fibroblasts from two female donors were provided by Dr. John Greenwood (Royal Adelaide Hospital, Adelaide, SA, Australia) and used between passages 2 and 4. Human mammary glandederived fibroblasts were provided by Dr. Wendy Ingman (The University of Adelaide, Adelaide, SA, Australia) and were isolated by enzymatic digestion of breast tissue,11 donated by five patients undergoing prophylactic mastectomies or cosmetic breast reduction surgery and used between passages 2 and 5. Human prostate stromal fibroblasts12 were provided by Dr. Damien Leach (The University of Adelaide). To evaluate the effect of peroxidases on collagen production, fibroblasts were seeded into 96-well plates at a density of 1.2  104 cells per well and cultured for 5 days in DMEM/10% FBS until reaching confluence. Cells were starved overnight in serum-free DMEM and then stimulated for an additional 72 hours in serum-free DMEM containing either ascorbic acid 2-phosphate at 100 mmol/L (Wako) as a Q6 positive control or with the various peroxidase proteins in the absence of ascorbic acid supplementation. At the end of the 72-hour treatment period, fibroblast-conditioned media were collected for measurement of secreted, soluble type I and VI collagen by enzyme-linked immunosorbent assay. Cell viability/growth was then assessed using the alamarBlue fluorescent dye assay (Invitrogen). Briefly, cells were incubated in a 10% alamarBlue/phosphate-buffered saline (PBS) solution for 30 minutes at 37 C and fluorescence measured at wavelengths of 530-nm excitation and 595-nm emission using a FLUOstar Optima plate reader (BMG Labtech). For time course studies, 6  104 HFFs Q7 were seeded into T25 culture flasks in DMEM/10% FBS and maintained in culture for 5 days. On reaching confluence, cells were starved in serum-free DMEM overnight and then stimulated with a maximal dose of 10 ng/mL transforming growth factor (TGF)-b2, 100 mmol/L ascorbic acid 2-phosphate, 1 mg/mL SBP, or 1.56 mg/mL recombinant human MPO in serum-free DMEM. Cell-conditioned medium and total RNA were harvested at time points up to and including 72 hours. For Western blot analyses of collagen I in HFF-conditioned medium, 6  105 HFFs were seeded into T75 culture flasks in DMEM/10% FBS and maintained in culture for 5 days. On reaching confluence, cells were starved in serum-free DMEM overnight and then stimulated in 8 mL serum-free DMEM containing either 100 mmol/L ascorbic acid 2-phosphate or 1.56 mg/mL peroxidases SBP, EPO, or MPO (for 72 hours).

Collagen I and VI Enzyme-Linked Immunosorbent Assay The amount of soluble type I and VI collagen in cellconditioned medium was measured by a direct coat enzymelinked immunosorbent assay method using standard curves constructed from purified type I and VI human placental collagen (BD Biosciences). Samples and standards (100 mL Q8

ajp.amjpathol.org

-

The American Journal of Pathology

FLA 5.2.0 DTD
AJPA1979_proof
7 March 2015
5:33 pm
EO: AJP14_0571

187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248

249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310

Peroxidases Regulate Collagen Q1 Q9

Q10

Q11

Q12

per well) were added to a 96-well Maxisorp plate (Nunc) and left at 4 C overnight. The plate was then washed with PBS-Tween 0.05%, 2.5% bovine serum albumin (BSA)/ PBS blocking solution was added to each well, and the plate was incubated for 1 hour at room temperature. The plate was then washed with PBS-Tween 0.05% and primary antibody (0.25 mg/mL rabbit anti-human collagen I polyclonal; 0.25 mg/mL rabbit anti-human collagen VI polyclonal; Rockland) in 5% nonfat dairy milk added to each well for 3 hours at room temperature. After washing, europium-tagged antirabbit secondary antibody (0.5 mg/mL in 1% BSA/PBS; Wallac Oy) was added for 1 hour at room temperature. After a final wash, Enhancement Solution (Wallac Oy) was added and fluorescence measured at excitation 355 nm and emission 620 nm using a FLUOstar Optima plate reader (BMG Labtech). The collagen content of each sample was determined from the standard curve (mg/mL), then normalized to DMEM-only treated cells.

311 to detect cellular distribution of both MPO and EPO. In 312 brief, cells were fixed with 4% formaldehyde with 5% su313 crose in PBS for 10 minutes. Cells were permeated with 314 Q18 0.25% Triton X-100 in PBS for 10 minutes to detect 315 intracellular peroxidase uptake; non-specific binding sites 316 were blocked by using 3% BSA in PBS containing 0.1% 317 glycine and a 1:10 dilution of nonimmune goat serum for 90 318 minutes. Cells were incubated for 1 hour with the primary 319 antibodies [1 mg/mL rabbit anti-human MPO from Dako Q19 320 321 (A0398); 1:1000 dilution of mouse anti-human EPO from Millipore (MAB1087)] in PBS containing BSA at 1 mg/mL Q20 Q21 322 323 and then counterstained with a 1:1000 dilution of Alexa Fluor 324 488econjugated goat anti-mouse or goat anti-rabbit IgG in 325 PBS containing 1 mg/mL BSA. Coverslips were mounted 326 onto glass slides using Fluoromount (Sigma-Aldrich), and 327 images were captured using a fluorescence photomicroscope 328 Q22 (Observer Z1; Zeiss). 329

RNA Isolation and Quantitative Real-Time PCR Western Blot Analysis

Q13 Q14

Q15

Q16 Q17

Culture medium from an equal number of HFFs grown in T75 flasks and stimulated for 72 hours in serum-free DMEM was resolved under reducing conditions in NuPAGE 4% to 12% Bis-Tris gels and electrophoretically transferred to Hybond-P polyvinylidene difluoride membranes (GE Healthcare) for 2 hours at 200 mA. The filters were blocked with blocking buffer (5% skim milk powder in Tris-buffered saline; 150 mmol/L NaCl, 50 mmol/L Tris, and 0.2% Tween 20, pH 7.4) overnight at 4 C and then incubated for 1 hour with 0.5 mg/mL rabbit anti-human collagen I polyclonal (Rockland). After extensive washing with several changes of Tris-buffered saline, the filters were incubated for 1 hour with alkaline phosphataseeconjugated goat anti-rabbit IgG (Thermo Scientific) at a 1:5000 dilution in blocking buffer. After washing, immunoreactivity was detected by using the ECF substrate reagent kit (GE Healthcare) on a FluorImager (Molecular Dynamics Inc.). The thermal stability of collagen I released by HFFs into the culture medium was assessed by digestion with a mixture of trypsin and chymotrypsin (Sigma-Aldrich) for 2 minutes at the various temperatures indicated.13 After digestion, a 20-mL aliquot was separated under reducing conditions in NuPAGE 4% to 12% Bis-Tris gels, and collagen I was analyzed by Western immunoblotting, as described above.

To evaluate the effect of peroxidases on collagen type Ia mRNA expression by HFFs, total RNA was isolated using an RNeasy Mini Kit (Qiagen) according to the manufac- Q23 turer’s instructions. RNA yield and purity were quantified by Nanodrop spectrophotometric measurement at 260 nm (Nanodrop Technologies). Collagen Ia mRNA expression Q24 was examined using real-time RT-PCR and normalized against glyceraldehyde-3-phosphate dehydrogenase. cDNA was synthesized by reverse transcription of 1 mg total RNA using random hexamer primers and SuperScript III Reverse Transcriptase (Invitrogen). Real-time RT-PCR was performed using SYBR Green Fluor qPCR mastermix (Qiagen) in a CFX96 Real-Time System (BioRad). Each reaction Q25 volume of 25 mL contained cDNA templates, primer pairs, and SYBR Green mastermix. Amplification occurred after initial denaturation at 95 C for 3 minutes, followed by 40 cycles at 95 C for 15 seconds, 60 C for 26 seconds, and 72 C for 10 seconds. The primer combination used for collagen I a1 was as follows: 50 -AGGGCTCCAACGAGATCGAGATCCG-30 (forward) and 50 -TACAGGAAGCAGACAGGGCCAACGTCG-30 (reverse). Expression values were normalized to the housekeeping gene glyceraldehyde3-phosphate dehydrogenase using the following primer combination: 50 -ACCCAGAAGACTGTGGATGG-30 (forward) and 50 -TCAGTGAGCTTCCCGTTCAG-30 (reverse).

Immunofluorescence Staining

Cell Migration Assay

HFFs (1.2  104 cells per well) were seeded onto circular glass coverslips in DMEM supplemented with 2 mmol/L glutamine, 100 IU/mL penicillin, 100 mg/mL streptomycin, 20 mmol/L HEPES, and 10% FBS (Invitrogen) and maintained in culture overnight. Cells were then treated with either saline (Control) or 1 mg of MPO or EPO in serum-free DMEM for up to 3 hours at 37 C. After peroxidase exposure, indirect immunofluorescence staining was performed

HFF migration was determined using the 24-well transwell plate (BD Falcon FluoroBlok) system with 8.0-mm pore Q26 positron emission tomography membranes. Briefly, HFFs were then seeded in serum-free DMEM at 10,000 cells per well into the upper wells. Cells were then incubated at 37 C for 30 minutes to allow cell adhesion. Lower chambers were then filled with 700 mL of serum-free DMEM as the vehicle control, DMEM containing 1 ng/mL TGF-b2 serving as the

The American Journal of Pathology

-

ajp.amjpathol.org

FLA 5.2.0 DTD
AJPA1979_proof
7 March 2015
5:33 pm
EO: AJP14_0571

3

330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372

373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434

DeNichilo et al positive control, or DMEM with the various peroxidase proteins (MPO, EPO, and SBP) at 2 mg/mL. After 20-hour incubation at 37 C, the inserts were gently taken out and the upper side of the membrane scraped using a cotton swab to remove cells that had attached but not migrated. Cells that had migrated to the underside of the transwell membrane were fixed in 6:1 ethanol/acetic acid for 10 minutes and stained with DAPI (1 mg/mL), then imaged and quantified on a fluorescent inverted microscope (Zeiss).

INTEGRA Studies Q27

Q28 Q29

Q30 Q31 Q32

Q33

Q34 Q35

INTEGRA Dermal Regeneration Template (INTEGRA LifeSciences) was cut into 1  1-cm pieces and washed with PBS, followed by serum-free DMEM to remove the alcohol storage medium and to equilibrate the porous scaffold for exposure to cells. The INTEGRA pieces were then transferred to a 24-well tissue culture plate (collagen layer facing up) and immersed in DMEM, with or without peroxidase, for up to 2 hours at 37 C in a CO2 incubator. After the preincubation binding period, the media were replaced with fresh DMEM supplemented with 5% human serum containing 5  105 human fibroblast cells, with or without SBP or HRP at 25 mg/mL. The INTEGRA pieces were incubated for up to 14 days, with the media changed after 7 days. At the end of the experiment, the INTEGRA pieces were imaged, harvested, and fixed in 10% buffered formaldehyde, processed, embedded in paraffin wax, and divided into sections for histological and immunohistochemical analysis. For immunohistochemistry, sections (6 mm thick) of the INTEGRA were mounted on silane-coated slides, dewaxed, and hydrated before placing into an endogenous peroxidase blocking solution for 30 minutes (0.5% H2O2/methanol). After washing in PBS, sections were incubated in 3% normal horse serum blocking solution for 30 minutes and then primary antibody overnight [2.5 mg/mL rabbit anti-smooth muscle actin (AbCam); 5 mg/mL rabbit anti-collagen I and collagen VI (Rocklands Immunochemicals)]. The sections were then washed in PBS and incubated with 6 mg/mL anti-rabbit IgG biotin. After a further PBS wash, 2 mg/mL streptavidin HRP tertiary antibody was applied for 1 hour, followed by the addition of 3,30 -diaminobenzidine peroxidase substrate solution (Sigma-Aldrich) to the sections until brown color development was visible. The sections were then washed thoroughly in PBS and water before counterstaining with hematoxylin, dehydrating, and mounting with DPX. Sections were also stained with DAPI (Pierce Biotechnology) to visualize cells within the INTEGRA template under fluorescent microscopy. DAPI was used at a final concentration of 1 mg/mL, following the manufacturer’s instructions, and cell counts were performed using an image analysis program (AnalySIS LS Starter; VisionWorks). Images were captured using a Nikon Eclipse 50i microscope attached to a DS-L2 control unit (Digital Sight) and a DS-Fi1 digital camera (Nikon Corporation).

4

In Vivo Rat Studies Female Sprague-Dawley rats (n Z 4) were lightly anesthetized, and the dorsum was shaved. At different sites on their dorsum, each rat received a bolus intradermal injection of four doses of peroxidase protein (SBP, 1.25 to 25 mg) or the vehicle control, which was 0.9% saline. Agents were injected in a volume of 50 mL, which was sufficient to generate a small blister in the skin that resolved within a few minutes. Across the group of rats, each dose was injected into a different injection site to control for any site-related, intrarat variation. After 3 days, the rats were sacrificed by CO2 asphyxiation and the skin at the injection sites was excised using a 6-mm biopsy punch, with each piece bisected and fixed in 10% buffered formalin. Fixed tissue was processed for histological assessment by graded dehydration and mounting in paraffin blocks. Sections (6 mm thick) were stained with hematoxylin and eosin (H&E) and Masson’s trichrome for qualitative analysis by an experienced pathologist. H&E-stained sections were Q36 scored for the fibroblast reaction (1þ to 3þ) to reflect the number of fibroblast cells identified at the injection site, with 3þ representing the greatest response. H&E- and Masson’s trichromeestained sections were used to evaluate the amount of deposited, collagen-rich ECM that was associated with the fibroblast reaction at each injection site. The ECM deposition was scored 1þ to 3þ, with 3þ representing the greatest response. The scores were combined across the group of rats, and statistical analysis was performed using a Kruskal-Wallis one-way analysis of variance on ranks, followed by a post hoc t-test comparing the response of the proteins with peroxidase activity to the control using Dunnett’s method. P < 0.05 was considered significant. All in vivo studies were approved by the Animal Ethics Committee of the Institute for Medical and Veterinary Sciences, and performed in accordance with the National Health Q37 and Medical Research Council of Australia code of practice for the care and use of animals for scientific purposes.

In Vivo Porcine Studies To determine the ability of peroxidase proteins to promote tissue regeneration in vivo, a study was performed in pigs (large white/Landrace cross sows) where INTEGRA (100 mg/mL SBP) was grafted onto full-thickness excisional wounds, as previously described.14 Two female domestic large white  Landrace pigs, weighing between 15 and 20 kg, obtained from The University of Adelaide Minimal Disease Piggery were used for this study. After induction of anesthesia by gaseous isoflurane, animals were intubated and deep anesthesia was maintained throughout the surgical intervention. The skin on the dorsum of the animal was rigorously prepared before generating four 3  3-cm square full-thickness wounds at two positions equidistant from the midline on either side.

ajp.amjpathol.org

-

The American Journal of Pathology

FLA 5.2.0 DTD
AJPA1979_proof
7 March 2015
5:33 pm
EO: AJP14_0571

435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496

497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558

Peroxidases Regulate Collagen

Mammalian and plant-derived peroxidase enzymes promote collagen I and VI release by human dermal fibroblasts. Enzyme-linked immunosorbent assay detection of soluble collagen I (A and C) and collagen VI (B and D) in human foreskin fibroblasts (HFFs) and adult dermal fibroblast-conditioned medium after 72 hours of stimulation with peroxidases at the doses indicated. Ascorbic acid 2-phosphate (AA) at 100 mmol/L served as the positive control, whereas the cells treated with Dulbecco’s modified Eagle’s medium (DMEM) alone (Unstim) served as the baseline control. The levels of collagen I and VI are expressed as fold change, normalized so the average values of Unstim were set to 1. Statistical significance was calculated by two-tailed Student’s t-test with the various treatment groups compared to the DMEM alone (Unstim) group. Data for adult dermal fibroblasts are pooled from experiments conducted using cells derived from two individual donors. Dose-response experiments using HFFs were independently performed at least 10 times, whereas studies using adult fibroblasts were repeated three times. Data are the means  SD of triplicate determinations for Unstim, AA, and each peroxidase dose. *P < 0.05, **P < 0.01, and ***P < 0.001. EPO, eosinophil peroxidase; HRP, horseradish peroxidase; MPO, myeloperoxidase; SBP, soybean peroxidase.

Figure 1

The site preparation involved removing the hair by clipping and application of a depilatory cream (Nair), followed by Q39 repeated scrubbing with MediSponges containing Betadine (povidone-iodine). The margins of the wounds were infiltrated Q40 with 0.25% bupivacaine (Marcaine) containing 1:400,000 adrenaline to facilitate hemostasis and afford postoperative analgesia. The skin was excised to the fascia layer and hemostasis effected by diathermy. INTEGRA pieces that had been soaked in a solution containing 100 mg/mL SBP or a saline solution (control) for 3 hours were placed into the wounds and secured in place with steel staples. The wounds were then covered by a moistened Acticoat dressing (Smith & Q41 Nephew), which was secured in place using Hypafix. The whole site was then protected using combine dressings underneath crepe bandages and an elasticized abdominal binder. The animals were dosed postoperatively with s.c. 250 mg Q42 Q43 Baytril 50 (Bayer Australia) as a prophylactic antibiotic. On Q38

The American Journal of Pathology

-

day 7, the animals were anesthetized and paired 6-mm punch biopsy specimens were collected from control and SBP-treated INTEGRA wounds. The biopsy specimens were fixed in 10% formalin, processed, and embedded in paraffin, and sections were stained with H&E and Masson’s trichrome for histological assessment. Images were captured using a Nikon Eclipse 50i microscope attached to a DS-L2 control unit (Digital Sight) and a DS-Fi1 digital camera (Nikon Corporation).

Results Peroxidases Stimulate Fibroblasts in Vitro to Secrete Collagens Type I and VI Type I collagen is the most abundant collagen found in the body, where it plays an important role in maintaining the

ajp.amjpathol.org

FLA 5.2.0 DTD
AJPA1979_proof
7 March 2015
5:33 pm
EO: AJP14_0571

5

Q46

559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620

DeNichilo et al 621 structural integrity of the major organs, including skin. When 622 stimulated with MPO or EPO in the absence of AA, HFFs 623 demonstrated a dose responsiveness to both mammalian 624 peroxidases. Like AA, which was used as a control, maximal 625 stimulation with the peroxidases increased the amount of 626 collagen I released into the medium approximately threefold 627 628 ½F1 (Figure 1A). To establish whether induction of collagen I secretion was restricted only to mammalian peroxidases, we 629 also tested peroxidase enzymes purified from plant sources. 630 631 Both SBP and HRP were potent stimulators of collagen 632 biosynthesis, achieving a maximal threefold increase in 633 collagen I release by HFFs, but at substantially lower doses 634 (Figure 1A). 635 Collagen VI is another form of collagen that plays an 636 important structural role in maintaining skin integrity, with a 637 recent report showing skin of Col6a1-null mice has 638 decreased tensile strength and abnormal collagen I fibril ar639 chitecture.15 Examination of soluble collagen VI synthesis by 640 HFFs revealed a similar release profile to that seen for 641 collagen I, with all peroxidases tested inducing a response 642 643 that was equal to, or greater than, that seen with AA stimu644 lation (Figure 1B). Throughout these experiments, the 645 peroxidase proteins had no impact on the proliferation or 646 viability of the confluent fibroblasts (data not shown). 647 To ensure the response to the peroxidase proteins was 648 not specific to neonatal fibroblasts, we tested the range 649 of peroxidase enzymes on fibroblasts derived from skin 650 removed during abdominoplasty from adult female donors. 651 Consistent with our observations made using HFFs, the 652 combined data from two adult donors showed that all four 653 peroxidases tested induced significant increases in collagen I 654 655 and VI secretion in the absence of AA (Figure 1, C and D). 656 Although the plant-derived peroxidases were again most 657 potent, both EPO and MPO stimulated levels of collagen I 658 and VI release that were comparable to the effect observed 659 with AA alone. Peroxidases showed no significant impact 660 on the viability or number of adult dermal fibroblasts (data 661 not shown). 662 The ability of peroxidase enzymes to stimulate collagen I 663 release in the absence of AA does not appear to be restricted 664 solely to fibroblasts sourced from dermal tissue. Combined 665 data from five donors show that cultured mammary fibro666 667 blasts also respond to both EPO and MPO and induce a 668 ½F2 significant increase in collagen I secretion (Figure 2A). 669 Similarly, a twofold increase in collagen I release was 670 observed when cultured human prostate fibroblasts were 671 stimulated with EPO or MPO in the absence of AA 672 (Figure 2B). 673 674 Peroxidases Promote Fibroblast Migration in Vitro 675 676 Fibroblast migration is central to many normal and patho677 logical processes involving tissue remodeling, where new 678 679 collagen-rich ECM is deposited, including wound repair, 680 fibrosis, cancer growth, and metastasis. To establish whether 681 peroxidase enzymes could provide a chemotactic stimulus 682

6

Figure 2 Eosinophil peroxidase (EPO) and myeloperoxidase (MPO) promote collagen I release by human mammarye and prostate-derived fibroblasts. Enzyme-linked immunosorbent assay detection of soluble collagen I in primary human mammary fibroblaste (A) and human prostate cancerassociated fibroblaste (B) conditioned medium after 72 hours of stimulation with EPO and MPO at the doses indicated. Ascorbic acid 2-phosphate (AA) at 100 mmol/L served as the positive control, whereas cells treated with Dulbecco’s modified Eagle’s medium (DMEM) alone (Unstim) served as the baseline control. The levels of collagen I are expressed as fold change, normalized so the average values of Unstim were set to 1. Statistical significance was calculated by two-tailed Student’s t-test with the various treatment groups compared to the DMEM alone (Unstim) group. Data for mammary fibroblasts are pooled from experiments conducted using cells derived from five individual donors. Data for the prostate cancereassociated fibroblasts are pooled from two separate experiments. Data are the means  SD of triplicate determinations for Unstim, AA, and each peroxidase dose. *P < 0.05, **P < 0.01, and ***P < 0.001.

to promote fibroblast migration in vitro, a transwell system was used where HFFs were seeded in the upper chamber with peroxidases present in the lower chamber acting as chemoattractants. During a 20-hour period, cell counts show the peroxidase enzymes MPO, EPO, and SBP significantly increased HFF migration approximately 2.5-fold compared

ajp.amjpathol.org

-

The American Journal of Pathology

FLA 5.2.0 DTD
AJPA1979_proof
7 March 2015
5:33 pm
EO: AJP14_0571

683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744

smooth muscle actinepositive cells appears to be due predominantly to the infiltration of fibroblasts into the site of SBP injection, rather than increased proliferation of resident cells. In addition, the response was no longer evident after 7 days, indicating remodeling had occurred, as would be expected of an acute-phase, fibroblastic response without the involvement of a chronic, inflammatory component.

Peroxidases Promote Tissue-Like Regeneration in Vitro We next sought to determine whether stimulation of fibroblasts with peroxidase enzymes could result in the generation of cell-associated collagen-rich ECM. For these studies,

Figure 3 Peroxidase enzymes promote fibroblast migration in vitro. Migration of human foreskin fibroblasts (HFFs) across transwell membranes after 20 hours of exposure to the peroxidase enzymes eosinophil peroxidase (EPO), myeloperoxidase (MPO), or soybean peroxidase (SBP) at 2 mg/mL in the lower chamber. Transforming growth factor (TGF)-b2 at 1 ng/mL in the lower chamber served as the positive control, whereas cells treated with Dulbecco’s modified Eagle’s medium (DMEM) alone (Unstim) in both chambers served as the baseline control. The level of cell migration is expressed as fold change, normalized so the average values of Unstim were set to 1. Statistical significance was calculated by two-tailed Student’s ttest with the various treatment groups compared to the DMEM alone (Unstim) group. Data are the means  SD of triplicate determinations for Unstim, TGF-b2, and each peroxidase. *P < 0.05, ***P < 0.001.

½F3

to the control (unstimulated), a response slightly greater to that achieved with the positive control TGF-b2 (Figure 3).

Peroxidases Stimulate a Fibrogenic Response When Injected in Vivo To confirm that peroxidase proteins could induce the same effects in vivo as those seen in vitro, SBP was injected intradermally into the skin of female rats. The results shown were ½F4 obtained 72 hours after SBP was injected (Figure 4, AeD). Compared to normal dermis (Figure 4A), the area injected with peroxidase protein showed a marked increase in cellularity and newly formed ECM (Figure 4C), which was confirmed by staining serial sections with Masson’s trichrome (data not shown). Although there was some lymphocytic cell involvement in the response to SBP, the cells were predominantly fibroblasts, as demonstrated by the strong positive Q44 staining for smooth muscle actin (Figure 4D). The blinded, histological evaluation of the injection sites from four rats injected with SBP showed that the peroxidase protein induced a dose-dependent fibrogenic response comprising an overall fibroblastic reaction (ie, increased fibroblast accumulation) and increased ECM deposition (Figure 4E). No apparent increase in the number of fibroblast mitotic bodies was evident, as confirmed by an absence of positive staining when the sections were incubated with an antibody against proliferating cell nuclear staining (data not shown). This accumulation of

The American Journal of Pathology

-

print & web 4C=FPO

745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806

Peroxidases Regulate Collagen

Peroxidases stimulate fibroblast migration and extracellular matrix (ECM) production in rats when injected intradermally. A and C: Representative images of hematoxylin and eosin (H&E)estained sections taken 3 days after rats were injected intradermally with either saline (control) or a 25-mg bolus dose of soybean peroxidase (SBP). B and D: Representative images of smooth muscle actin (SMA) immunostaining showing fibroblast distribution in tissue sections 3 days after intradermal injection with either saline (control) or a 25-mg bolus dose of SBP. E: Histological evaluation of H&E- and Masson’s trichromeestained sections measuring overall tissue fibroblast reaction and extracellular matrix (ECM) deposition 3 days after various doses of SBP were injected intraderamlly into rats. Statistical significance was calculated using a Kruskal-Wallis oneway analysis of variance on ranks, followed by a post hoc t-test comparing the response of the proteins with peroxidase activity to the control (vehicle) using Dunnett’s method. The intradermal injection of peroxidase enzymes into rats was independently performed three times. Data are the means  SEM (E). n Z 4 rats per group (E). *P < 0.05, **P < 0.01. Scale bar Z 250 mm (B and D).

Figure 4

ajp.amjpathol.org

FLA 5.2.0 DTD
AJPA1979_proof
7 March 2015
5:33 pm
EO: AJP14_0571

7

Q47

807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868

DeNichilo et al

print & web 4C=FPO

869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 we used INTEGRA Dermal Regeneration Template as a 897 premade three-dimensional collagen/glycosaminoglycan 898 scaffold that was seeded with adult donor fibroblasts. The 899 plant peroxidases SBP and HRP were chosen for these 900 studies, because these enzymes stimulate the most potent 901 collagen response of all peroxidases tested in adult fibroblast 902 903 monolayer cultures (Figure 1, C and D). After 7 days, we 904 noted a consistent difference in the ability of cells to infiltrate 905 into INTEGRA when treated with peroxidases compared to 906 culture medium alone. This was illustrated by DAPI staining, 907 where INTEGRA slices treated with SBP showed elevated 908 numbers of fibroblasts extending throughout the scaffold, 909 indicating that the cells had either attached themselves to the 910 collagen fibers of the scaffold or migrated into the interstices 911 between the fibers (Figure 5A). In comparison, media alone ½F5 912 were not effective at promoting cell migration into the 913 INTEGRA (Figure 5A). Cell counts compiled from DAPI914 915 stained sections demonstrated that both SBP and HRP 916 effectively doubled the number of cells populating the 917 INTEGRA after 7 days (Figure 5B). 918 The ability of cells that had migrated into the INTEGRA 919 scaffold to deposit a collagen-rich ECM in response to the 920 treatment with peroxidase proteins was assessed by immu921 nostaining for collagen I using a polyclonal antibody that 922 also cross-reacts with bovine collagen I. As a result, the 923 fibers making up the INTEGRA stained positive in all im924 ages shown for collagen I staining (Figure 5, C and D). Cell/ 925 INTEGRA composites, generated using cells from adult 926 927 donors and cultured without peroxidase protein for 2 weeks, 928 showed predominantly this endogenous staining pattern 929 with minimal cell-associated collagen Iepositive staining 930

8

Figure 5 Peroxidases promote fibroblast migration and collagen extracellular matrix (ECM) deposition in vitro. A: Representative images of DAPI-stained INTEGRA dermal regeneration template showing adult dermal fibroblast localization after 7 days of culture without peroxidase (control) or stimulated with soybean peroxidase at 25 mg/mL (SBP). B: Cell counts of DAPI-stained adult dermal fibroblasts within INTEGRA sections after 7 days of culture without peroxidase (control), stimulated with soybean peroxidase at 25 mg/mL (SBP) or horseradish peroxidase at 25 mg/mL (HRP). Pooled data from five experiments. Statistical significance was calculated by two-tailed Student’s ttest. C: Representative images of INTEGRA harvested after 14 days of culture with adult dermal fibroblasts maintained without peroxidase (control) or stimulated with soybean peroxidase at 25 mg/mL (SBP) showing both collagen I and collagen VI distribution by immunostaining. D: Representative images of INTEGRA harvested after 14 days of culture with adult dermal fibroblasts maintained without peroxidase (control) or stimulated with horseradish peroxidase at 25 mg/mL (HRP) showing both collagen I and collagen VI distribution by immunostaining. Each in vitro experiment using INTEGRA was independently performed at least five times. Data are the means  SEM (B). *P < 0.05 compared to control (B). Scale bars: 400 mm (A); 250 mm (D).

Q48

evident (Figure 5, C and D). In contrast, when SBP or HRP was included in the media, collagen I staining was observed throughout the interstices of the INTEGRA, clearly indicating the peroxidase proteins stimulated the robust deposition of collagen Ieenriched ECM (Figure 5, C and D). The collagen VI antibody did not cross-react with the collagen I component of the INTEGRA scaffold, and minimal collagen VI was deposited within the interstices of the scaffold when the cell/INTEGRA composite was cultured without peroxidase proteins (Figure 5, C and D). Conversely, when SBP or HRP was included in the media, profuse deposition of collagen VI was observed throughout the matrix (Figure 5, C and D).

Peroxidases Induce Connective Tissue Regeneration in Vivo To determine whether an in vivo fibrogenic response could be elucidated within the INTEGRA scaffold by the peroxidase proteins, INTEGRA presoaked with normal saline (control) or with SBP for 3 hours was grafted onto full-thickness dermal wounds on the dorsum of pigs (Figure 6, A and B). Histo- ½F6 logical analysis of biopsy specimens collected 7 days after the INTEGRA grafting showed strong evidence of advanced integration of the peroxidase-treated INTEGRA with the underlying wound bed (Figure 6D). Closer inspection shows that the control INTEGRA was poorly infiltrated with cells from the wound bed (Figure 6, C and E), resulting in poor attachment and the separation of the INTEGRA from the wound tissue during collection. In contrast, the INTEGRA treated with SBP was well integrated with new granulation

ajp.amjpathol.org

-

The American Journal of Pathology

FLA 5.2.0 DTD
AJPA1979_proof
7 March 2015
5:33 pm
EO: AJP14_0571

931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992

Peroxidases Regulate Collagen amount of collagen I stimulated by these enzymes increased with time, reaching maximal levels between 48 and 72 hours. The profile of collagen I release stimulated by the peroxidases was virtually identical to that observed for AA, in terms of both magnitude and time dependency. In contrast, without AA supplementation, TGF-b2 failed to induce a significant increase in collagen I release over the 72-hour time frame

print & web 4C=FPO

993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 Q49 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 ½F7 1054

Figure 6 Peroxidase enzymes stimulate potent tissue regeneration in vivo when combined with INTEGRA dermal regeneration template. A and B: The wounds with INTEGRA stapled in place (control and SBP treated, respectively) before biopsy collection at 7 days after surgery. C and D: Representative low-power images of sections stained with Masson’s trichrome, clearly showing that only the INTEGRA itself is harvested from the control wound (C), whereas INTEGRA and the underlying wound tissue are recovered from the SBP wound (D). E and F: Representative highermagnification images showing the control INTEGRA is poorly infiltrated with cells from the wound bed (E), whereas the INTEGRA treated with SBP (F) shows numerous fibroblasts and inflammatory cells interspersed with new extracellular matrix deposited between the INTEGRA fibers at the wound bed interface. The pig excisional wound model using INTEGRA was independently performed on two pigs with similar results. Scale bars: 600 mm (C and D); 120 mm (E and F). Original magnifications: 40 (C and D); 200 (E and F).

tissue from the wound (Figure 6F), showing numerous fibroblasts and new collagen-rich ECM interspersed between the fibers of the INTEGRA at the interface with the wound bed, along with several inflammatory cells.

Peroxidases Regulate P4H-Dependent Post-Translational Collagen I Biosynthesis To investigate the mechanism by which peroxidase enzymes regulate collagen I biosynthesis, we compared the effect of MPO and SBP to the time-dependent secretion of collagen I stimulated by either AA or TGF-b2. In the absence of AA, both MPO and SBP increased the levels of collagen I detected in HFF-conditioned media after just 6 hours (Figure 7A). The

The American Journal of Pathology

-

Figure 7

Peroxidases induce collagen I release without regulating collagen I a1 mRNA levels. A: Enzyme-linked immunosorbent assay detection of soluble collagen I levels in human foreskin fibroblast (HFF)econditioned medium at the indicated time points on stimulation with 1.56 mg/mL myeloperoxidase (MPO), 1 mg/mL soybean peroxidase (SBP), 100 mmol/L ascorbic acid 2-phosphate (AA), or 10 ng/mL transforming growth factor (TGF)-b2. Cells treated with Dulbecco’s modified Eagle’s medium (DMEM) alone (Unstim) served as the baseline control at each time point. The levels of soluble collagen I are expressed as mg/mL. B: Quantitative real-time PCR analysis of the collagen I a1 mRNA expression (normalized to glyceraldehyde-3phosphate dehydrogenase) in HFFs at the indicated time points on stimulation with 1.56 mg/mL MPO, 1 mg/mL SBP, 100 mmol/L AA, or 10 ng/mL TGF-b2. Cells treated with DMEM alone (Unstim) served as the baseline control at each time point. The levels of collagen I a1 transcript are expressed as fold change, normalized so the average values of Unstim at each time point were set to 1. Statistical significance was calculated by two-tailed Student’s t-test, with the various treatment groups compared to the DMEM alone (Unstim) group at each time point. Each time course experiment was independently performed three times. Data are the means  SD of triplicate determinations for each time point. *P < 0.05, **P < 0.01, and ***P < 0.001.

ajp.amjpathol.org

FLA 5.2.0 DTD
AJPA1979_proof
7 March 2015
5:33 pm
EO: AJP14_0571

9

1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116

1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 ½F8 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178

DeNichilo et al (Figure 7A). Real-time PCR analysis of collagen I a1 mRNA levels within these cells revealed that the robust increase in collagen I protein stimulated by the peroxidases was not associated with a corresponding increase in collagen I mRNA levels (Figure 7B). In contrast, TGF-b2 clearly induced a time-dependent increase in collagen I a1 mRNA, with a maximal threefold increase detected at 24 hours, with levels returning to baseline by 72 hours. Compared to TGF-b, AA induced a small, but significant, increase in collagen Ia mRNA expression between 6 and 48 hours after stimulation, which is consistent with its established role as a profibrogenic factor that predominantly regulates collagen biosynthesis at a post-translational level.4,16 AA acts on cultured fibroblasts to promote efficient hydroxylation of procollagen, leading to the subsequent release of triple helical collagen from the cell that is thermally stable by nature.2 Our findings that peroxidases are capable of stimulating collagen I release by fibroblasts in a reaction that does not require AA suggest these enzymes are also likely to influence post-translational hydroxylation of procollagen I. It was, therefore, important to investigate whether collagen I released by peroxidase-treated cells in the absence of AA was similar biochemically to the collagen I released by fibroblasts stimulated with AA alone. We performed Western blot analysis of collagen I released by HFFs into the culture medium because gross differences in hydroxylation can be detected as changes in procollagen subunit mobility during gel electrophoresis.17 The data show both the bands corresponding to the collagen proa1(I) and proa2(I) subunits migrate at the same molecular weight irrespective of whether the cells were stimulated with AA alone or with peroxidase enzymes in the absence of AA (Figure 8A), indicating that no significant differences exist in the level of collagen hydroxylation. To assess the thermal stability of collagen I released by fibroblasts on peroxidase treatment in the absence of AA, culture medium from HFFs stimulated with either AA or EPO was subjected to digestion using a mixture of trypsin and chymotrypsin over an increasing temperature range between 34 C and 42 C, followed by gel electrophoresis and Western blot analysis for collagen I. The results clearly show that treatment of HFFs with either AA or EPO results in the release of collagen I with an identical Tm (midpoint of thermal transition) of 40 C (Figure 8B), suggesting that the collagen I released by peroxidase-treated cells in the absence of AA is fully or near fully hydroxylated and triple helical. As the AA and peroxidase-mediated collagen I release profiles, Tm and electrophoretic mobilities were similar; further in vitro studies using cultured fibroblasts were conducted to establish whether peroxidases were also capable of regulating post-translational collagen I biosynthesis in a P4H-dependent manner. HFFs were stimulated with either MPOorEPOinthepresenceofdimethyloxalylglycine(DMOG), a cell-permeable structural analogue of a-ketoglutarate, which acts as a competitive inhibitor of prolyl hydroxylases.18

10

Figure 8 Western blot analysis of collagen I secreted by human foreskin fibroblasts (HFFs) after exposure to peroxidase enzymes or ascorbic acid 2-phosphate (AA). A: Western blot analysis of collagen I from culture medium of HFFs stimulated 72 hours with 1.56 mg/mL peroxidase [myeloperoxidase (MPO), eosinophil peroxidase (EPO), or soybean peroxidase (SBP)], 100 mmol/L AA, or Dulbecco’s modified Eagle’s medium (DMEM) alone (Unstim). Bands corresponding to proa1(I) and proa2(I) and molecular weights (kDa) are indicated. B: Western blot analysis of collagen I from culture medium of HFFs stimulated 72 hours with either 100 mmol/L AA or 1.56 mg/mL EPO and subsequently digested with a mixture of trypsin and chymotrypsin for 2 minutes at temperatures indicated.

As a positive control, cells stimulated with AA were also treated with DMOG to confirm if the release of collagen I was dependent on P4H-mediated hydroxylation. DMOG blocked the AA and also the peroxidase-induced release of collagen I from HFFs in a dose-dependent manner, with complete inhibition occurring between 25 and 50 mmol/L (Figure 9). The inhibitory effect of DMOG was not associated with a loss in cell viability (data not shown).

Peroxidase Catalytic Activity Is Essential for Collagen I Biosynthesis When HFFs were stimulated with MPO or EPO in the presence of 4-amino-benzoic acid hydrazide (4-ABAH), an irreversible inhibitor of peroxidase enzymatic activity,19 a clear dose-dependent inhibition of collagen I release was observed over a 72-hour time period (Figure 10). In contrast, 4-ABAH over the same dose range had no impact on AAinduced collagen I release, confirming 4-ABAH was inhibiting the peroxidase catalytic activity and not the cellular collagen biosynthetic pathway. HFF exposure to increasing

ajp.amjpathol.org

-

The American Journal of Pathology

FLA 5.2.0 DTD
AJPA1979_proof
7 March 2015
5:33 pm
EO: AJP14_0571

1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 ½F9 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 ½F10: 1235 1236 1237 1238 1239 1240

1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 Q50 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 ½F11: 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302

Peroxidases Regulate Collagen

Figure 9 Inhibition of peroxidase-induced collagen I release by the prolyl hydroxylase inhibitor dimethyloxalylglycine (DMOG). Enzyme-linked immunosorbent assay detection of soluble collagen I levels in human foreskin fibroblast (HFF)econditioned medium after cells were pretreated for 30 minutes with DMOG at the doses indicated, then stimulated by the addition of 5 mg/mL myeloperoxidase (MPO), 5 mg/mL eosinophil peroxidase (EPO), or 100 mmol/L ascorbic acid 2-phosphate (AA) for a further 72 hours. Soluble collagen I levels are expressed as fold change, normalized to the average values of cells treated with Dulbecco’s modified Eagle’s medium alone (Unstim). Statistical significance was calculated using the two-tailed Student’s t-test with the DMOG-treated groups compared to the treatment group without DMOG. Inhibition studies using DMOG were independently performed twice over the same dose range. Data are the means  SD of triplicate determinations. *P < 0.05, **P < 0.01, and ***P < 0.001.

migration and collagen I biosynthesis, offering a new and important insight into the causative association between inflammation and aberrant fibrotic conditions. For example, the intense deposition of EPO (originating from degranulating eosinophils) found only in the fibrotic connective tissue of human endometriosis specimens and not in normal proliferative or secretory endometrium directly implicates this peroxidase in the fibrotic sequel associated with this condition.9 Similarly, in the light of the results presented herein, MPO is significantly increased in bronchoalveolar lavage fluid taken from individuals with idiopathic pulmonary fibrosis20,21; however, elevated levels of both EPO and MPO have been detected in sputum of patients with chronic obstructive pulmonary disease.22 Moreover, Rudolph et al8 have provided clinical and experimental evidence, using MPO-deficient mice, that MPO acts as a profibrotic mediator in atrial fibrillation, although these authors postulated that the molecular mechanism underlying MPO-dependent atrial fibrosis was via regulation of matrix metalloproteinase enzymatic activity, rather than the regulation of collagen biosynthesis. Peroxidase-mediated collagen biosynthesis by tumorassociated fibroblasts may also play an important role in tumor progression and metastasis. Several reports have recently emerged showing MPO localization and enhanced catalytic activity to be associated with the early-stage development of both lung and breast tumors in mouse models,23e25 whereas 88% of human breast cancer biopsy specimens showed EPO deposits located primarily in the connective

doses of 4-ABAH up to 250 mmol/L had no impact on cellular viability over the duration of the experimental period (data not shown). These data confirm that the hemecontaining catalytic domain plays an essential role in MPOand EPO-mediated collagen I biosynthesis in fibroblasts.

Fibroblasts Rapidly Bind and Internalize Peroxidase Enzymes To determine whether exposure of HFFs to MPO and EPO could promote cell surface binding and internalization of these enzymes, immunofluorescence localization studies were performed (Figure 11). The images clearly show that cultured dermal fibroblasts bind and rapidly internalize peroxidase enzymes. Both MPO and EPO accumulate on the cell surface within minutes of peroxidase exposure (Figure 11), whereas no staining was detected in peroxidase-untreated cells, confirming the specificity of the EPO and MPO immunostaining (data not shown). After 3 hours of peroxidase exposure, there was intense intracellular MPO and EPO staining, with both enzymes localizing to vesicle-like structures that have a distinct perinuclear distribution (Figure 11).

Discussion Our data provide evidence, for the first time, that the peroxidase enzymes EPO and MPO regulate fibroblast

The American Journal of Pathology

-

Figure 10

Inhibition of peroxidase-induced collagen I release by 4amino-benzoic acid hydrazide (4-ABAH). Enzyme-linked immunosorbent assay detection of soluble collagen I levels in human foreskin fibroblast (HFF)econditioned medium after cells were pretreated for 30 minutes with 4-ABAH at the doses indicated, then stimulated by the addition of 5 mg/mL myeloperoxidase (MPO), 5 mg/mL eosinophil peroxidase (EPO), or 100 mmol/L ascorbic acid 2-phosphate (AA) for a further 72 hours. Soluble collagen I levels are expressed as fold change, normalized to the average values of cells treated with Dulbecco’s modified Eagle’s medium alone (Unstim). Statistical significance was calculated using the two-tailed Student’s t-test with the 4-ABAHetreated groups compared to the treatment group without 4-ABAH. Inhibition studies using 4-ABAH were independently performed three times. Data are the means  SD of triplicate determinations. *P < 0.05, **P < 0.01, and ***P < 0.001.

ajp.amjpathol.org

FLA 5.2.0 DTD
AJPA1979_proof
7 March 2015
5:33 pm
EO: AJP14_0571

11

Q51

1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364

1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426

DeNichilo et al

Figure 11 Localization of peroxidase distribution by indirect immunofluorescence. Cultured human foreskin fibroblasts were exposed to 1 mg/mL myeloperoxidase (MPO) or eosinophil peroxidase (EPO) in serumfree Dulbecco’s modified Eagle’s medium for either 20 minutes (surface) or 3 hours (intracellular). Cells were fixed, permeabilized (intracellular only), and stained for MPO and EPO localization, as described in Materials and Methods. Original magnification: 40 (surface images); 63 (intracellular images).

tissue stroma around and within the tumor.26 Our discovery that MPO and EPO can also regulate mammary and prostate stromal fibroblast collagen I biosynthesis strengthens the case that these enzymes potentially serve as therapeutic targets to prevent the rapid escalation in collagen I biosynthesis seen in pathological diseases, such as cancer, where inflammatory cell infiltration and peroxidase release are key features. Our experimental data indicate that peroxidase enzymes do not stimulate fibroblast proliferation either under culture conditions or when injected intradermally into rats. These observations are contrary to those recently reported where MPO stimulated an increase in synovial fibroblast proliferation, as measured by intracellular bromodeoxyuridine incorporation.27 An important feature that may account for the discrepancy between our observations and those reported in synovial fibroblasts is that tumor necrosis factor a, a well-recognized fibroblastic mitogenic factor,28 was also added in combination with MPO to promote proliferation of these cells. Whether MPO alone has the capacity to stimulate synovial fibroblast proliferation was not assessed. Our mechanistic data from in vitro culture studies demonstrate that the peroxidase proteins induce a cellular response that is clearly distinct from TGF-b in terms of collagen mRNA and protein expression. In contrast, the overall response elicited by the peroxidase proteins is not dissimilar to those generally reported for ascorbate. Our finding that exposure of fibroblasts to peroxidase proteins caused a robust increase in collagen I secretion without a corresponding increase in collagen I mRNA levels is consistent with previous findings reported after stimulation of human skin fibroblasts with ascorbate.16 AA is known to function as an enzymatic cofactor for prolyl hydroxylase to promote post-translational collagen biosynthesis.2e4 Our

12

data showing the inhibition of peroxidase-induced collagen I release by the prolyl hydroxylase inhibitor DMOG support our hypothesis that post-translational events are most likely to be affected by the peroxidases. Moreover, given that hydroxylation is an essential requirement for secretion of procollagen from the cell, we propose that peroxidase enzymes can substitute for AA at this point in the biosynthetic pathway to improve hydroxylation efficiency and the use of translated protein. Our results showing that in the absence of AA, collagen I released by peroxidase-treated fibroblasts has a Tm of 40 C add further support to this notion, indicating the collagen I is hydroxylated and in a thermally stable triple helical conformation. We believe this important observation has previously been overlooked by others because AA is routinely added to fibroblast culture medium when examining collagen biosynthesis. When combined with AA, the stimulatory effects of peroxidase enzymes on fibroblast collagen release are masked (data not shown), suggesting that both agents target the same regulatory point in collagen biosynthesis. Our data also confirm that peroxidase heme-containing catalytic domain plays an essential role in MPO- and EPOmediated collagen biosynthesis in fibroblasts. Both mammalian and plant-derived peroxidase enzymes consume hydrogen peroxide (H2O2) as a substrate to generate multiple distinct intracellular reactive oxidant species.29 Cellular H2O2 is predominantly produced through dismutation of superoxide that is formed by the one-electron reduction of molecular oxygen in a reaction catalyzed by NADPH oxidase.30 H2O2 is lipid soluble, can readily diffuse across cellular membranes, and is known to oxidize the Fe2þ bound to the prolyl hydroxylase enzyme to Fe3þ via the Fenton reaction, leading to inhibition of prolyl hydroxylase enzymatic activity and substrate hydroxylation.31 Our observation that cultured fibroblasts rapidly bind and internalize both MPO and EPO raises the distinct possibility that peroxidase enzymes regulate the reduction-oxidationesensitive prolyl hydroxylase system by catalytically depleting cellular H2O2 levels. Although the precise mechanisms are unclear, it is conceivable that intracellular reactive oxidant species generated by peroxidases also contribute to collagen biosynthesis by acting as alternative electron donors to reactivate prolyl hydroxylase iron to the Fe2þ state in the absence of AA. Most known reactive species generated by peroxidases are short lived and produced at low levels, making the task of identifying the exact oxidant species capable of regulating P4H enzymatic activity and collagen biosynthesis a difficult one. In conclusion, our novel findings, coupled with those previously reported by others, suggest that the peroxidase Q45 enzymes MPO and EPO are likely to regulate multiple cellular processes involved in collagen ECM regeneration, including fibroblast recruitment, collagen biosynthesis, and modulation of matrix metalloproteinase activity. The convergent outcome of these multiple mechanisms strongly suggests that these enzymes play an important role not only in normal tissue repair, but also in the progression of disease

ajp.amjpathol.org

-

The American Journal of Pathology

FLA 5.2.0 DTD
AJPA1979_proof
7 March 2015
5:33 pm
EO: AJP14_0571

1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488

1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551

Peroxidases Regulate Collagen states where infiltrating inflammatory cells are known to deposit peroxidases.

Acknowledgments Dr. Pritinder Kaur (Peter Mac Callum Institute, Melbourne, VIC, Australia) provided human foreskin fibroblasts; Dr. John Greenwood (Royal Adelaide Hospital, Adelaide, SA, Australia) provided adult dermal fibroblasts; Dr. Wendy Ingman (The University of Adelaide, Adelaide, SA, Australia) provided human mammary glandederived fibroblasts; and Dr. Damien Leach (The University of Adelaide) provided human prostate stromal fibroblasts.

References 1. Peterkofsky B: Ascorbate requirement for hydroxylation and secretion of procollagen: relationship to inhibition of collagen synthesis in scurvy. Am J Clin Nutr 1991, 54:1135Se1140S 2. Myllyharju J, Kivirikko KI: Collagens and collagen-related diseases. Ann Med 2001, 33:7e21 3. Canty EG, Kadler KE: Procollagen trafficking, processing and fibrillogenesis. J Cell Sci 2005, 118:1341e1353 4. Kivirikko KI, Myllylä R, Pihlajaniemi T: Protein hydroxylation: prolyl 4-hydroxylase, an enzyme with four cosubstrates and a multifunctional subunit. FASEB J 1989, 3:1609e1617 5. Parsons KK, Maeda N, Yamauchi M, Banes AJ, Koller BH: Ascorbic acid-independent synthesis of collagen in mice. Am J Physiol Endocrinol Metab 2006, 290:E1131eE1139 6. Nishida H, Kurahashi T, Saito Y, Otsuki N, Kwon M, Ohtake H, Yamakawa M, Yamada K, Miyata S, Tomita Y, Fujii J: Kidney fibrosis is independent of the amount of ascorbic acid in mice with unilateral ureteral obstruction. Free Radic Res 2014, 48:1115e1124 7. Klebanoff SJ: Myeloperoxidase: friend or foe. J Leukoc Biol 2005, 77: 598e625 8. Rudolph V, Andrie RP, Rudolph TK, Friedrichs K, Klinke A, HirschHoffmann B, Schwoerer AP, Lau D, Fu X, Klingel K, Sydow K, Didié M, Seniuk A, von Leitner E, Szoecs K, Schrickel JW, Treede H, Wenzel U, Lewalter T, Nickenig G, Zimmermann W, Meinertz T, Böger RH, Reichenspurner H, Freeman BA, Eschenhagan T, Ehmke H, Hazen SL, Willems S, Baldus S: Myeloperoxidase acts as a profibrotic mediator of atrial fibrillation. Nat Med 2010, 16:470e474 9. Blumenthal RD, Samoszuk M, Taylor AP, Brown G, Alisauskas R, Goldenberg DM: Degranulating eosinophils in human endometriosis. Am J Pathol 2000, 156:1581e1588 10. Forbes E, Murase T, Yang M, Matthaei KI, Lee JJ, Lee NA, Foster PS, Hogan SP: Immunopathogenesis of experimental ulcerative colitis is mediated by eosinophil peroxidase. J Immunol 2004, 172:5664e5675 11. Stingl J, Emerman JT, Eaves CJ: Enzymatic dissociation and culture of normal human mammary tissue to detect progenitor activity. Methods Mol Biol 2005, 290:249e263 12. Li Y, Li CX, Ye H, Chen F, Melamed J, Peng Y, Liu J, Wang Z, Tsou HC, Wei J, Walden P, Garabedian MJ, Lee P: Decrease in stromal androgen receptor associates with androgen-independent disease and promotes prostate cancer cell proliferation and invasion. J Cell Mol Med 2008, 12:2790e2798 13. Bruckner P, Prockop D: Proteolytic enzymes as probes for the triplehelical conformation of procollagen. Anal Biochem 1981, 110: 360e368

The American Journal of Pathology

-

14. Greenwood JE, Dearman BL: Comparison of a sealed, polymer foam biodegradable temporizing matrix against Integra dermal regeneration template in a porcine wound model. J Burn Care Res 2012, 33:163e173 15. Lettmann S, Bloch W, Maab T, Niehoff A, Schulz JN, Eckes B, Eming SA, Bonaldo P, Paulsson M, Wagener R: Col6a1 null mice as a model to study skin phenotypes in patients with collagen VI related myopathies: expression of classical and novel collagen VI variants during wound healing. PLoS One 2014, 9:1e9 16. Chan D, Lamande SR, Cole WG, Bateman JF: Regulation of procollagen synthesis and processing during ascorbate-induced extracellular matrix accumulation in vitro. Biochem J 1990, 269:175e181 17. Cheah KS, Grant ME, Jackson DS: Translation of type II procollagen mRNA and hydroxylation of the cell-free product. Biochem Biophys Res Commun 1979, 91:1025e1031 18. Baader E, Tschank G, Baringhaus K, Burghard H, Günzler V: Inhibition of prolyl 4-hydroxylase by oxalyl amino acid derivatives in vitro, in isolated microsomes and in embryonic chicken tissues. Biochem J 1994, 300:525e530 19. Kettle AJ, Gedye CA, Winterbourn CC: Mechanism of inactivation of myeloperoxidase by 4-aminobenzoic acid hydrazide. Biochem J 1997, 321:503e508 20. Hällgren R, Bjermer L, Lundgren R, Venge P: The eosinophil component of the alveolitis in idiopathic pulmonary fibrosis: signs of eosinophil activation in the lung are related to impaired lung function. Am Rev Respir Dis 1989, 139:373e377 21. Stijn W, Verleden SE, Vanaudenaerde BM, Wynants M, Dooms C, Yserbyt J, Somers J, Verbeken EK, Verleden GM, Wuyts WA: Multiplex protein profiling of bronchoalveolar lavage in idiopathic pulmonary fibrosis and hypersensitivity pneumonitis. Ann Thorac Med 2013, 8:38e45 22. Keatings VA, Barnes PJ: Granulocyte activation markers in induced sputum: comparison between chronic obstructive pulmonary disease, asthma, and normal subjects. Am J Respir Crit Care Med 1997, 155: 449e453 23. Zhang N, Francis KP, Prakash A, Ansaldi D: Enhanced detection of myeloperoxidase activity in deep tissue through luminescent excitation of near-infrared nanoparticles. Nat Med 2013, 19:500e550 24. Alshetaiwi HS, Balivada S, Shresha TB, Pyle M, Basel MT, Bossmann SH, Troyer DL: Luminol-based bioluminescence imaging of mouse mammary tumors. J Photochem Photobiol B 2013, 127:223e228 25. Rymaszewski AL, Tate E, Yimbesalu JP, Gelman AE, Jarzembowski JA, Zhang H, Pritchard KA, Vikis HG: The role of neutrophil myeloperoxidase in models of lung tumor development. Cancers (Basel) 2014, 6: 1111e1127 26. Samoszuk MK, Nguyen V, Gluzman I, Pham JH: Occult deposition of eosinophil peroxidase in a subset of human breast carcinomas. Am J Pathol 1996, 148:701e706 27. Odobasic D, Yang Y, Muljadi RCM, O’Sullivan KM, Kao W, Smith M, Morand EF, Holdsworth SR: Endogenous myeloperoxidase is a mediator of joint inflammation and damage in experimental arthritis. Arthritis Rheumatol 2014, 66:907e917 28. Thornton SC, Por SB, Penny R, Richter M, Shelley L, Breit SN: Identification of the major fibroblast growth factors released spontaneously in inflammatory arthritis as platelet derived growth factor and tumour necrosis factor-alpha. Clin Exp Immunol 1991, 86:79e86 29. Davies MJ, Hawkins CL, Pattison DI, Rees MD: Mammalian heme peroxidases: from molecular mechanisms to health implications. Antioxid Redox Signal 2008, 10:1199e1234 30. Genestra M: Oxyl radicals, redox-sensitive signalling cascades and antioxidants. Cell Signal 2007, 19:1807e1819 31. Gerald D, Berra E, Frapart YM, Chan DA, Giaccia AJ, Mansuy D, Pouyssegur J, Yaniv M, Mechta-Grigoriou F: JunD reduces tumor angiogenesis by protecting cells from oxidative stress. Cell 2004, 118: 781e794

ajp.amjpathol.org

FLA 5.2.0 DTD
AJPA1979_proof
7 March 2015
5:33 pm
EO: AJP14_0571

13

1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614