ADVERTISEMENT
The Effect of Collagenase on Ischemic Wound Healing: Results of an In Vivo Study
Abstract
Many chronic wounds have a limited blood supply and contain necrotic tissue that must be debrided. The effect of collagenase, an enzymatic debriding agent, has been studied in acute but not in chronic wounds. The purpose of this in vivo study is to evaluate the effect of collagenase on wound healing in ischemic wounds.
The ears of eight young New Zealand White rabbits were rendered ischemic by ligation of caudal and central arteries and dermal circulation circumferentially so both ears were perfused only by the rostral artery, preserving the caudal, central, and rostra veins. Three 6-mm, full-thickness dermal punches were made on the inner surface of both ears down to perichondrium. One ear on each rabbit was treated with either collagenase or petrolatum ointment covered with a semi-occlusive dressing; wounds on the other ear of the same rabbit were covered with a semi-occlusive dressing only (control).
On post-wounding day 8, wound samples were collected and processed for histological analysis of reepithelialization (epithelial gap, percentage healed, epithelial height, and epidermal area) and granulation tissue formation (peak height, granulation tissue distance, and area). Within-animal comparison showed no significant differences between the petrolatum and control wounds but epithelial height, epidermal area, wound peak height, and granulation tissue distance and area were significantly different between the collagenase and control-treated wounds. Between-animal comparison of petrolatum- and collagenase-covered wounds showed statistically significant (P <0.05) differences for the following outcomes: epithelial gap, percenatge healed, epithelial height, epidermal area, wound peak height, and granulation tissue distance and area. In this ischemic wound model, outcomes for most of the variables associated with healing were significantly better in wounds covered with collagenase and a semi-occlusive dressing than in the control or petrolatum group. Additional in vivo studies are warranted to confirm these results.
Potential Conflicts of Interest: Dr. Jia, Dr. Law, and Dr. Mustoe disclose they are employees of Healthpoint Biotherapeutics, Ltd, Fort Worth, TX. Dr. Galiano discloses he has received research grants from Healthpoint Biotherapeutics, Ltd, Sanuwave (Alpharetta, GA), Smith & Nephew (St. Petersburg, FL), and Excaliard Pharmaceuticals (San Diego, CA).
Wounds that have failed to return to functional and anatomical integrity in a timely fashion are defined as chronic.1,2 Pressure ulcers, diabetic ulcers, and venous ulcers comprise the overwhelming majority of chronic wounds.2 These ulcer types share common causative features: the cellular and systemic changes of aging, repeated ischemia-reperfusion injury, and bacterial colonization with a resulting inflammatory host response.3 Although the etiologies of all chronic wounds are multifactorial, ischemia—ie, a decrease in blood flow—and/or ischemia-reperfusion injury appears to be a primary causality. One of the major reasons for impaired wound healing is an attenuated blood supply, which results in a decrease in the concentration of growth factors within the wound bed.2 This is particularly true for lower extremity wounds and/or pressure ulcers.
Collagenases have been used as an enzyme-debriding agent for chronic dermal ulcers and severe burn wounds. Although collagenases act by degrading native helical collagen fibrils,4 preliminary results from animal studies show that collagenase-treated, partial-thickness wounds in pigs healed quicker than control-treated wounds; furthermore, significant pro-angiogenic activity was observed in mice with basement membrane matrix implants (Matrigel™, BD, Franklin Lakes, NJ) (unpublished data). The effects of collagenase in acute wound healing has been studied in a swine back, acute, full-thickness wound model, but not for chronic wounds.5 Although many animal models are available, only a small number of these models actually mimic chronic wounds; one of these is the unique impaired wound healing model caused by ischemia in a rabbit ear.6
Because information about the potential effect of collagenase on chronic wound healing is limited, the purpose of this in vivo study was to evaluate the effect of collagenase on granulation tissue formation and reepithelialization under ischemic conditions.
Materials and Methods
Eight (8) young adult New Zealand White rabbits (3 to 6 months, ~3 kg; Covance Research Products, Inc., Cumberland, VA) were acclimated, housed, and given access to food and water ad libitum under the conditions set forth in the Public Health Service Guide for the Care and Use of Laboratory Animals and the US Department of Agriculture Title 9–Animal Welfare Act and its revisions. The study was conducted under an approved experimental protocol from the Northwestern University Animal Care and Use Committee.
Both ears in each of the eight rabbits were rendered ischemic by the method described by Ahn et al.6 Briefly, the rabbit was anesthetized with an intramuscular (IM) injection of ketamine (45 mg/kg) and xylazine (7 mg/kg) and prophylactically treated with penicillin (50,000 units/kg, IM) before surgery. The surgical sites and the dorsal surface of the rabbit ears were shaved with an electric clipper and depilated with Nair (Church & Dwight Co., Inc., Lakewood, NJ). The surgical sites then were painted with a betadine solution and an incision was made to the level of bare cartilage at the base of the ear. Under a dissecting microscope, the caudal and central arteries, as well as dermal circumferential circulation, were ligated so the entire ear was perfused only by the rostral artery with preservation of the caudal, central, and rostral veins to render the rabbit ear ischemic (an important reason for delayed healing). This technique results in ischemia for 7 to 10 days. In aged rabbits with sustained ischemia, no healing has been known to occur up to 26 days.7 No truly chronic wounds have been noted in animals, but this model has been extensively characterized and used to test therapeutic options for clinical use.6
The incision was closed with a 5-0 polypropylene suture. Three 6-mm, full-thickness dermal punches then were made on the inner surface of both ears down to bare cartilage by removing the epidermis, dermis, and perichondrium. One ear of the rabbit served as a nontreated control ear covered by a transparent semi-occlusive dressing consisting of a thin polyurethane membrane coated with a layer of an acrylic adhesive (Tegaderm™ dressing, 3M, St. Paul, MN). The wounds in the other ear of the same rabbit were treated with 0.05 mL of either collagenase ointment (Santyl® Ointment, Healthpoint Biotherapeutics, Ltd, Forth Worth, TX—approximately 1.23 g collagenase in a petrolatum-based ointment) or petrolatum alone, and covered by the same semi-occlusive dressing. The same agent (either petrolatum or collagenase) was used on all three wounds (1 cm to 1.5 cm apart) on the treated ear. Each rabbit received one treatment in one ear and no treatment in the other ear. All animals were sacrificed at post-wounding day 8.
Tissue from all the wounds was collected and analyzed. The tissue of each wound was equally split into two halves. One half was processed for histological analysis and the other half was stored at -80º C or colder for future western blot analysis or RNA extraction.
The collagenase and petrolatum test products were provided by Healthpoint Biotherapeutics, Ltd (Fort Worth, TX) in individually sealed, coded, and labeled aluminum tubes. The tubes were stored at room temperature.
Histological analysis. The tissue from half of the circular wound taken for histological analysis was subject to routine paraffin embedding and sectioning. A 4-µm cross-section was obtained through the center of each circular wound. Tissues were stained with hemotoxylin and eosin (H&E) for microscopic examination and measurement. Histological analyses included reepithelialization in terms of the epithelial gap (EG) and the percentage healed (PH); keratinocyte proliferation in terms of the epithelial height (EH) and the epidermal area (EA); and the formation of granulation tissue in terms of the peak height, granulation tissue distance, granulation tissue area, and the granulation tissue area per µm wound (see Table 1). All measurements were performed using a Nikon microscope and accompanying imaging software at 2X and 10X magnification.
Because the measured wound size among slides was not identical due to differences in cutting sections or slightly off-center in some samples, the ratio of 6 mm (the original wound size) to the measured wound size was used to adjust the raw data, with the exception of PH and epidermal area per µm wound and granulation tissue area per µm wound measurements, which already were adjusted.
Data and statistical analysis. There were 12 wounds for each of the collagenase- and petrolatum-treated groups and their respective controls. To detect a difference between the collagenase or petrolatum treatment and its own control groups, the data were compared using two-tailed paired t-tests (SAS® Institute Inc., Cary, NC). Analysis of variance (ANOVA) with Tukey’s test was used to compare outcomes of the collagenase or petrolatum treatment groups from different animals while controlling for differences between animals. P <.05 was considered significant in all analyses.
Results
Wound reepithelialization. No differences in epithelial gap and percentage healed were observed between the collagenase-treated ischemic and nontreated control wounds (EG: 1,123 ± 389 µm versus 1,779 ± 497 µm, P = .4; PH: .81 ± .06 versus .70 ± .08, P = .4). No significant difference in reepithelialization measures was found between petrolatum-treated and nontreated controls (EG: 3,858 ± 754 µm versus 2,707 ± 583 µm, P = .3; PH: .36 ± .13 versus .55 ± .10, P = .3). However, between-animal comparison of collagenase- and petrolatum-treated wounds showed significant differences (EG collagenase 1,123 ± 389 µm versus 3,858 ± 754 µm petrolatum group, P = .004 and PH collagenase 81% ± .06 versus 36% ± .13 petrolatum group, P = .004) (see Figures 1 and 2). Keratinocyte proliferation, as measured by epithelial thickness and area, was significantly higher in collagenase-treated wounds than either of the respective control (untreated ischemic ear) or the petrolatum-treated wounds (EH: 165 ± 11 µm versus 117 ± 8 µm or 79 ± 18 µm, P = .004 or <.0001; EA: 152 ± 6 µm2 versus 98 ± 8 µm2 or 66 ± 16 µm2, P <.0001 or <.0001). However, no significant difference was noted in keratinocyte proliferation between petrolatum-treated wounds and their own nontreated controls (EH: 79 ± 18 µm versus 124 ± 15 µm, P = .10; EA: 66 ± 16 µm2 versus 100 ± 10 µm2, P = .10) (see Figures 3 and 4).
Granulation tissue formation. Average tissue peak height, granulation tissue distance, and tissue area were higher in the collagenase than in the collagenase-control and petrolatum-treated wounds. Peak height measured 684 ± 28 µm versus 535 ± 42 µm or 442 ± 56 µm, P = .002 or .0006. Granulation tissue distance measured 1,451 ± 110 µm versus 989 ± 111 µm or 713 ± 170 µm, P = .02 or .0002. Granulation tissue area was 790,907 ± 118,010 µm2 versus 407,288 ± 85,867 µm2 or 285,596 ± 127,739 µm2, P = .007 or .003. Granulation tissue area per µm wound was 132 ± 20 µm2 versus 68 ± 14 µm2 or 48 ± 21 µm2, P = .007 or .003). Yet no significant difference was noted regarding granulation tissue formation between petrolatum-treated wounds and their own nontreated controls (peak height: 442 ± 56 µm versus 451 ± 40 µm, P = .90; granulation tissue distance: 713 ±170 µm versus 852 ± 60 µm, P = .50; granulation tissue area: 285,596 ± 127,739 µm2 versus 317,127 ± 62,178 µm2, P = .80; and granulation tissue area per µm wound: 48 ± 21 µm2 versus 53 ± 10 µm2, P = .80) (see Figures 5 through 8).
Discussion
Pierce and Mustoe8 described the basic principles of optimal wound healing to include minimizing tissue damage, débriding nonviable tissue, maximizing tissue perfusion and oxygenation, ensuring proper nutrition, and providing a moist wound-healing environment. Most preclinical study designs control certain conditions, such as minimizing tissue damage (by exclusion from study) and proper nutrition. Collagenase is an enzymatic debridement agent and although this study did not focus on maintaining a moist wound environment, all wounds (including control) were covered with a semi-occlusive dressing. The observed increased rate of reepithelialization in the collagenase-treated compared to the petrolatum-treated wounds is similar to that observed in another study,5 although the control in the latter study is dry gauze. However, following clinical and laboratory studies, Ghadially et al19 found that applying petrolatum did not completely occlude the epidermis of acetone-treated skin in both humans and mice; the authors concluded that petrolatum permeates throughout the epidermis interstices to allow recovery of the skin barrier function. Hence, petrolatum affects wound healing despite being viewed as inert.20
In the present study, collagenase ointment was significantly more effective compared to treatment with petrolatum for the following wound healing variables assessed: reduction in wound epithelial gap, percentage healed, epithelial height, epidermal area, peak tissue height, granulation tissue distance, and granulation tissue area. When compared with their own respective control wounds (an ischemic ear with no ointment applied), no significant differences were observed between the petrolatum and petrolatum control wounds; whereas, epithelial height, epidermal area, peak height, granulation tissue distance, granulation tissue area, and granulation tissue area per µm wound were significantly different between the collagenase and collagenase-control wounds. Differences between collagenase and control wounds in epithelial gap and percentage healed were not significant. Additional in vivo studies including direct petrolatum control or using other wound healing models are needed to confirm these results.
Conclusion
In this ischemic wound model, wounds treated with a topical debriding agent (collagenase) covered with a semi-occlusive dressing had better wound healing-related outcomes than wounds covered with a semi-occlusive dressing alone (untreated control) or covered with petrolatum. Additional preclinical studies are warranted and clinical studies to confirm the potential effects of collagenase ointment on healing chronic wounds are needed.
Acknowledgment
The authors are grateful to Healthpoint Biotherapeutics, Ltd, Fort Worth, TX, for its support and acknowledge Renée Carstens for medical writing contributions.
Dr. Jia and Dr. Zhao are research associates, The Laboratory for Wound Repair and Regenerative Medicine, Division of Plastic Surgery, Department of Surgery, Northwestern University, Chicago, IL. Dr. Law is Principal Research Scientist/Toxicology Professional, Healthpoint Biotherapeutics, Ltd, Fort Worth, TX. Dr. Galiano is an assistant professor and Dr. Mustoe is a professor, Division of Plastic Surgery, Northwestern University. Please address correspondence to: Thomas A. Mustoe, The Laboratory for Wound Repair and Regenerative Medicine, Division of Plastic Surgery, Department of Surgery, Northwestern University, 4th Floor, Tarry, Suite 4-720, 303 East Chicago Avenue, Chicago, IL 60611-3093; email: tmustoe@nmh.org.
1. Telgenhoff D, Shroot B. Cellular senescence mechanisms in chronic wound healing. Cell Death Differ. 2005;12(7):695–698.
2. Eaglstein WH, Falanga V. Chronic wounds. In: Barbul A (ed). The Surgical Clinics in North America: Wound Healing. Philadelphia, PA: WB Saunders Co.;1997:689–700.
3. Mustoe TA. Understanding chronic wounds: a unifying hypothesis on their pathogenesis and implications for therapy. Am J Surg. 2004;187(5A):65S–70S.
4. Collagenase, Clostridium histolyticum collagenase (E.C. 3.4.24.3) In: Worthington V (ed). Worthington Enzyme Manual, Enzymes and Related Biochemicals. Lakewood, NJ: Worthington Biochemical Corporation;1993:112–120.
5. Riley KN, Herman IM. Collagenase promotes the cellular responses to injury and wound healing in vivo. J Burn Wound. 2005;4:112–124.
6. Ahn ST, Mustoe TA. Effects of ischemia on ulcer wound healing: a new model in the rabbit ear. Ann Plast Surg. 1990;24(1):17–23.
7. Bonomo SR, Davidson J, Tyrone JW, Lin X, Mustoe TA. Enhancement of wound healing by hyperbaric oxygen and transforming growth factor-3 in a new chronic wound model in aged rabbits. Arch Surg. 2005;135(10):1148–1153.
8. Pierce GF, Mustoe TA. Pharmacologic enhancement of wound healing. Ann Rev Med. 1995;46:467–481.
9. Ayello EA, Cuddigan JE. Debridement: controlling the necrotic cellular burden. Adv Skin Wound Care. 2004;17(2):66–75.
10. Bradley M, Cullum N, Sheldon T. The debridement of chronic wounds: a systematic review. Health Technol Assess. 1999;3(17 Pt 1):iii–78.
11. Brookens T. Secrets of wound debridement yield improved outcomes and incomes. Podiatr Manage. 1997(June):101–102.
12. Falabella AF. Debridement and wound bed preparation. Dermatol Ther. 2006;19(6):317–325.
13. Ramundo J, Gray M. Enzymatic wound debridement. J WOCN. 2008;35(3):273–280.
14. Steed DL. Debridement. Am J Surg. 2004;187(5A):71S–74S.
15. Adams SB Jr, Sabesan VJ, Easley ME. Wound healing agents. Foot Ankle Clin. 2006;11(4):745–751.
16. Eaglstein WH, Davis SC, Mehle AL, Mertz PM. Optimal use of an occlusive dressing to enhance healing. Arch Dermatol. 1998;124(3):392–395.
17. Eaglstein WH, Mertz PM, Falanga V. Occlusive dressings. Am Fam Phys. 1987;35(3):211–216.
18. Eaglstein WH. Effect of occlusive dressings on wound healing. Clinics Dermatol. 1984;2(3):107–111.
19. Ghadially R, Halkier-Sorensen L, Elias PM. Effects of petrolatum on stratum corneum structure and function. J Am Acad Dermatol. 2001;26(3 Pt 2):387–396.
20. Eaglstein WH, Mertz PM. “Inert” vehicles do affect wound healing. J Investig Dermatol. 1980;74(2):90–91.
