- Email: [email protected]
Hyaluronan induces scarless repair in mouse limb organ culture
Hyaluronan
Induces By Joseph
A. locono,
Scarless Repair H. Paul
Ehrlich,
Hershey,
Kerry
Methods: Time-dated pregnant CD-l mice (term, 20 days) were killed on gestational day 18 and fetuses were harvested via laparotomy. A through and through stab wound was made in each forelimb with a l-mm microscapel, and the wound was closed with a single IO-O nylon suture. The forelimbs were amputated at the level of the shoulder and placed in organ culture. Daily medium changes with 1 mL of BGJb (devoid of serum) were made. Half the cultures received 10 uL of HA (4 mg/mL) directly to the wound site with each medium change. The other half of the cultures received 10 uL of phosphate-buffered saline (PBS-control). At day 7, the limbs were harvested, fixed in methyl Carnoys solution, paraffin embedded, and 5-pm serial sections cut. The sections were stained with H&E or Sirius red/fast green. The sections were viewed in a blinded fashion by two observers. Suture defined the wound site, and the sections were graded for healing by scarring.
CAR FORMATION terminates a cascade of events that is catalyzed by an inflammatory response that includes the release of cytokines, hormones, and growth factors. Platelets, neutrophils, and macrophages release these factors within the early wound milieu. The degree to which each component is required in the tissue repair process is unclear. Moreover, alterations in these elements are assumed to play a role in scarless fetal repair. Repair in organ cultured fetal forelimb explants limits the environment to local fibroblasts and their surrounding extracellular matrix. The model facilitates the study of From the Section of Pediatric Surgery, Department of Surgery, Milton S. Hershey h4edical Center; Hershey, PA. This work was supported through grants GM 41343 (TMK) and GM Fl7566 (JAI) from the National Institutes of Health. Address reprint requests to Joseph A. Iocono, MD, Section of Pediatric Surgery, Department of Surgery, Milton S. Hershey Medical Centet; The Penn State G&singer Health System, 500 University Di; Hershey, PA 17033-0850. Copyright o 1998 by W.B. Saunders Company 0022.3468/98/3304-0004$03.00/O
564
A. Keefer,
and
Thomas
Organ M.
Culture
Krummel
Pennsylvania
Background/Purpose:Wounded fetal mouse limbs harvested from two distinct time points in gestation heal differently in organ culture. The healing of a gestational day 14 limb is by scarless repair, whereas gestational day 18 (gd 18) limbs heal by scarring. The persistence of elevated levels of hyaluronic acid (HA) is a major difference in the extracellular matrix of scarless repair. The purpose of this study was to demonstrate that chronic additions of HA to incisional wounds of gd 18 limbs induces scarless repair.
S
in Mouse Limb
Results: Minimal limb growth occurred in both control and HA-treated limbs. Grossly, both control and treated limbs healed incisional wounds by 7 days in culture. Limbs from both treatment and control groups showed viability by microscopic analysis. The limbs treated with HA had no appreciable scar morphologically in sections in which epithelial dimpling and suture were evident. The orientation of the collagen fiber bundles in the control wounds were in parallel arrays perpendicular to the incision. The orientation of the collagen fiber bundles in the HA-treated limbs had a basket weave pattern that was indistinguishable from unwounded dermis. The direct repeated additions of HA to healing organ cultured limb explants of gestational day 18 fetal mice promoted scarless repair. Conc/usions:This result demonstrates that chronic elevation of HA in the microenvironment of a wound affects healing by promoting the deposition of a more dermal-like connective tissue matrix in the wound site. The maintenance of elevated levels of HA could have utility in the clinical setting to improve the organization of connective tissue, leading to the reduction of scar complications. J Pediatr Surg 33:564-567. Copyright o 1998 by W.B. Saunders Company. INDEX culture,
WORDS: scarless
Hyaluronic healing.
acid,
fetal
tissue
repair,
organ
single factors to the wound site and eliminates contributions from the circulation. The isolated, unperfused, fetal mouse limb demonstrates that skin wound repair proceeds in the absence of a circulation. Mouse forelimb organ cultures of gestational day 14 (term, 20 days) heal suture-closed incisional wounds without scar, whereas fetal mouse forelimbs isolated from gestational day 18 healed identical wounds with scar. Fetal mouse forelimb explants remained viable, grew,r and differentiated2 in organ culture. Moreover, the system was shown to be sensitive to added factors such as transforming growth factor beta (TGF-P), in which it promoted repair by scarring in the 14-day limb.3 The effect of daily addition of hyaluronic acid (HA) to healing l&day fetal mouse limbs in organ culture was studied. MATERIALS
AND
METHODS
Mice Time-dated pregnant CD-1 mice (Charles River Laboratories, Wilmington, MA) were individually caged and fed a standard rodent chow and water ad libitum. On gestation day 18 (term, 20). the pregnant
~ourna/ofPediatric
Surgery,
Vol33,
No 4 (April),
1998: pp 564-567
HYALURONAN
INDUCES
SCARLESS
REPAIR
females were killed by cervical dislocation, then hysterectomies performed. The entire uterus with fetuses was placed in ice cold phosphate-buffered saline (PBS). Each fetus was individually removed from its amniotic sac and placed in a petri dish under a dissecting microscope. A linear, l-mm full-thickness incision wound was made in each forelimb. The wound was closed with a single 10-O nylon suture at the midpoint of the incision. Each limb was then amputated at the level of the shoulder and placed in an organ culture dish. The experiment was run in triplicate (n = 12 for each treatment group).
Organ Culture System Organ culture dishes (#3037, Falcon Plastics, Lincoln, NJ), 60 mm X 15 mm designed with a central well and surrounding moat were placed into standard lo-cm petri dishes. The outer moat was filled with sterile water to maintain a humid environment. Stainless steel, wire mesh triangles (Industrial Wire Products, Pasadena, CA) were sterilized, then placed over the central well. A permeable 0.2~pm filter was then placed on top of the grid. BGJ-b medium (Gibco, Gaithersburg, MD), a chemically defined culture medium was supplemented with penicillin G, 1000 U/mL; streptomycin, 0.1 mg/mL; amphotericin B, 0.25 mg/mL; and ascorbate, 1 mg/mL @H adjusted to 7.4 to 7.5 with 1.0 mmol/L NaOH). The central well was filled with 1 mL of the medium, which was even with the surface of the wire mesh triangle. Half the cultures (n = 12) received 10 pL of PBS (control). whereas the other half (n = 12) received 10 pL HA (Sigma, St Louis, MO) to a final concentration of 0.4 mg/mL. The limbs were placed on the membrane, which was placed on the stainless steel mesh, the dish covered and incubated at 37°C in humidified 95% air with 5% CO2 for 7 days. The medium was changed daily.
Documentation of Gross and Histological Changes Gross morphological changes in the limbs were documented by photography on day 7 before harvesting. Histological examination was performed at 7 days by fixing the limbs in methyl Camoys fixative (10% acetic acid, 60% methanol, and 30% chloroform, v/v) for 24 hours, then replacing it with absolute ethanol and stored at -20°C until processing. Sections were embedded in pat&in, 5-pm serial sections were cut through the entire wound and then stained with either H&E or Sirius red with Fast green. Sections were viewed with both bright field and polarized light, and photographs were taken with Kodak Ektachrome 200 ASA slide film. Viability of each limb was confirmed with H&E-stained sections. Collagen bundles were easily visualized by their birefringence pattern exhibited under polarized light. Limbs were scored by two independent observers for wound closure and presence of scarring. Wound was defined in sections with visible suture and epidermal dimpling present. Scarring was defined by the presence of thick parallel collagen bundles beneath the epidermis under polarized light. Absence of scarring was defined by the presence of normal dermal architecture in the wound area and absence of thick parallel bundles of collagen.
565
were reviewed per limb. As expected, there was no evidence of inflammatory response in these organ cultured limbs. In the control group, all 12 limbs showed increased collagen deposition in the healed wound, consistent with tissue repair with scar. When examined under polarized light, the collagen fiber bundles were thicker than surrounding unwounded dermis and in parallel arrays (Fig 1). However, examination of the healed wound of the 12 limbs treated with HA uniformly showed normal dermal and epidermal architecture. When viewed with polarized light, the collagen fiber bundles deposited were in a basket weave pattern, and there was less collagen deposited compared with controls (Fig 2). Little difference in collagen fiber pattern was evident between wound and surrounding dermis. The chronic additions of HA to gd 18 incisional wounds in fetal limbs promoted scarless repair. DISCUSSION
The organ culture explant system facilitates the study of individual variables in the wound healing process within a complete intact organ. The fetal limb contains an organized dermis and epidermis as well as developing muscle and bone. Individual factors can be evaluated with this system. The fetal limb explant is dependent only on bathing culture medium for its nutrients. Several influences, extrinsic to the local wound environment, possibly play a role in regulating tissue repair. The degree of maturation in the fetal immune system and its inability to mount an inflammatory response to injury has been investigated.4 Further, the topical effects of different
RESULTS
In both the PBS-control and HA groups, wound closure was evident in all specimens by gross and histological inspection. Limb length did not appreciably increase, but further differentiation of digits and claw growth were evident (consistent with previous results). Histological differences in the character of the tissue repair between the control and HA-treated groups were found. There were 12 wounded limbs reviewed from each group. An average of 15 slides (20 sections per slide)
Fig 1. A representative cross section of a wounded forelimb from the control group stained with Sirius red as viewed with polarized light. The wound is shown between the arrows. Note the collagen fiber bundle architecture with thick parallel arrays (scar) beneath a regenerated epidermal layer. (Original magnification x400.)
566
Fig 2. A representative cross section of a wounded forelimb from the HA treatment group, stained with Sirius red as viewed with polarized light. The wound is shown between the arrows. The collagen fiber bundle architecture is the same as unwounded dermis. There is no appreciable scar under the regenerated epidermis. (Original magnification x400.)
growth factors found in the amniotic fluid are also under close study.5 They may change as fetal development continues until their increased concentrations may promote scar formation. The organ culture system has the ability to investigate any of these factors alone or in combination. Isolated limb cultures grown in a serumfree medium succeed in eliminating extrinsic factors of tissue repair. It is an unperfused system in which circulating cellular elements are absent. The culture media contains only defined materials where serum and amniotic fluid are excluded. HA is a negatively charged glycosaminoglycan composed of repeated disaccharides of D-glucuronic acid and N-acetylglucosamine.6 HA is the major glycosaminoglycan of the fetal extracellular matrix.7 HA has been shown to influence a number of cellular processes, which include tumor metastasis4 and inflammation8 HA also has been shown to enhance the healing of cornea1 wounds.9J0 Abantangelo et al” demonstrated that, in diabetic rats, HA enrichment of open wounds promoted quicker healing. Alexander and Donoff reported that HA initially accumulated in adult rabbit open wounds but was reduced by the appearance of hyaluronidase (HAdase).12 In contrast, the induction of HAdase was not seen in fetal wounds that healed without scar.13 Scar tissue is defined by the presence of thick bundles in a parallel array beneath a regenerated epidermis. These thick bundles seen in scar are the result of the deposition and organization of collagen fibers. Collagen fiber bundle deposition in a
IOCONO
ET AL
basket weave pattern indistinguishable from the pattern of unwounded dermis defines scarless repair. The amount and organization of collagen fiber bundles can be studied using Sirius red stained serial sections of wounds under polarized light. This technique demonstrates the density of collagen bundles by their intensity and birefringence pattern. As an example, the collagen fibers of normal dermis exhibit a basket weave birefringence pattern, whereas scar collagen fibers are thicker and are parallel to the epidermal layer. Grading of serial sections for collagen intensity and organization of the collagen bundles has been a useful tool for semiquantitative assessment of scarring in 1%day fetal mouse limbs.14 This technique was used to study the effect of chronic HA addition on repair in organ culture. The spatial organization of collagen fiber bundles within connective tissue can be demonstrated by Sirius red staining and polarized light viewing.15 Examination of the color and intensity of the birefringence pattern of collagen fiber bundles provides a visible means for documenting the degree of collagen organization. Immature collagen fiber bundles show fine fibrils having a green birefringence. Advanced, more organized collagen fiber bundles are thicker with a yellow-orange birefringence in a basket weave pattern. Young granulation tissue shows fine collagen fiber bundles in parallel arrays with green birefringence, and more mature granulation tissue or scar shows thick collagen fiber bundles arranged in parallel arrays with yellow-orange birefringence.16 There has been some debate as to whether the maintenance of high levels of HA in wounds that heal scarlessly is a causative factor or epiphenomenon of scarless fetal repair.17 This study shows that HA by itself evokes scarless repair. The fetal extracellular matrix must facilitate the growth, differentiation, and organogenesis of the developing fetus. Fetal matrix is rich in HA.‘* As fetal growth and differentiation nears completion at the end of gestation, HA levels decline.r3 Postnatally, during the normal repair mechanism, HA levels are initially elevated. However, in the adult wound, HA is degraded, whereas in fetal wounds, HA remains elevated.19 Midgestation fetal rabbit wounds, which repair scarlessly, can be induced to heal with scar by treating them with Hyaluronidase (HAdase), which specifically degrades HA.20 The mechanisms whereby HAdase eliminates scarless repair remains unclear. Reducing the levels of HA may affect cell signaling through modulation of the cell surface receptors.21-23 The mechanism for regenerative repair in the 14-day fetal limb explant and the adultlike repair in l&day limb explants is unclear.’ It is proposed that differences in GAG content of the extracellular matrix of the fetal dermis is responsible for these differences in the repair
HYALURONAN
INDUCES
SCARLESS
567
REPAIR
process. The fetal mouse limb organ culture system agrees with the in vivo system of fetal sheep and rabbitsz4 Scarless fetal tissue repair occurs in the intact fetus during early and midgestation (14-day mouse fetal limb). Later in gestation scarless repair is replaced with adult scar repair (1 g-day fetal limb). However, because of its complexity, the in vivo system is difficult to assess mechanistically. The well-defined fetal mouse limb explant organ culture system allows the reduction in the complexity of the system and offers a simpler system for
the study of mechanisms responsible for scarless fetal repair. Maintaining elevated levels of HA in adult wounds may decrease or eliminate scarring. The elucidation of the exact cellular mechanisms whereby HA exerts its effects may be helpful in designing therapeutic interventions to control scarring. The ability to maintain a sustained level of HA for a defined period of time within a wound site may reduce the complications of scarring in the future.
REFERENCES 1. Bleacher JC, Adolph VR, Dillon PW, et al: Isolated fetal mouse limbs: Gestational effects on tissue repair in an unperfused System. J Pediatr Surg 28(1):1312-1315, 1993 2. Chopra V, Blewett CJ, Ehrlich HP, et al: The transition from fetal to adult repair occurring in forelimbs maintained in organ culture. Wound Rep Regen 5:47-51,1997 3. Meisler N, Keefer KA, Ehrlich HP, et al: Dexamethasone abrogates the fibrogenic effect of transforming growth factor-in rat granuloma and granulation tissue fibroblasts. J Invest Derm 108:285-289, 1997 4. Knudson CB, Knudson W: Hyaluronan-binding proteins in development, tissue homeostasis, and disease. FASEBJ 7:1233-1241, 1993 5. Puolakkainen PA. Twardzik DR, Ranchalis JE, et al: The enhancement in wound healing by transforming growth factor-beta 1 TGF-beta 1) depends on the topical delivery system. J Surg Res 58:321-329, 1995 6. Iozzo RV: Proteoglycans: Structure, function and role in neoplasia. Lab Invest 53:373-396, 1985 7. DePalma RL, Krummel TM, Durham LA, et al: Characterization and quantification of wound matrix in the fetal rabbit. Matrix 9:224231,1989 8. Weigel PH, Fuller GM, Le Boeuf RD: A model for the role of Hyaluronic acid and fibrin in the early events during the inflammatory response and wound healing. JTheor Biol119:219-234,1986 9. Nakamura M, Nishida T, Hikida M, et al: Combined effects of hyaluronan and fibronectin on cornea1 wound closure of rabbit in vivo. Curr Eye Res 13:385-388,1994 10. Nishida T, Nakamura M, Mishima H, et al: Hyaluronan stimulates cornea1 epithelial migration Exp Eye Res 53:753-758, 1991 11. Abatangelo G, Martelli M, Vecchis P: Healing of hyaluronic acid enriched wounds: Histological observations. J Surg Res 35:410-416, 1983 12. Alexander SA, Donoff B: The glycosaminoglycans of open wounds. J Surg Res 29:422-426, 1980
13. Krummel TM, Nelson JM, Dieglemann RF, et al: Fetal response to injury in the rabbit. J Pediatr Surg 22:640-644, 1987 14. Cooney RN, Iocono JA, Maish G, et al: Tumor necrosis factor mediates impaired wound healing in chronic abdominal sepsis. J Trauma 42:415-420, 1997 15. Constantine FS, Mowry RW: The selective staining of human dermal collagen: The use of picrosirus red with polarization microscopy. J Invest Derm 50:419-423,1968 16. Ehrlich HP, Desmouliere A, Diegelmann RF: Morphological and immunochemical differences between keloid and hypertrophic scar. Am JPathol 145:105-113, 1994 17. Ferguson MW, Whitby DJ, Shah M, et al: Scar formation: The spectral nature of fetal and adult wound repair. Plast Reconstr Surg 97854-860, 1996 18. Depalma RL, Krummel TM, Nelsom et al: Fetal wound matrix is composed of proteoglycan rather than collagen. Surg Forum 38:626628, 1987 19. Estes JM, Adzick NS, Harrison MR, et al: Hyaluronate metabolism undergoes an ontogenic transition during fetal development: Implications for scar-free healing. J Pediatr Surg 28:1227-1231, 1993 20. Mast BA, Haynes JH, Krummel TM, et al: In vivo degradation of fetal wound hyaluronic acid results in increased fibroplasia, collagen deposition and neovascularization. Plast Reconstr Surg. 89:503-509, 1992 21. Toole BP: Hyaluronan and its binding proteins, the hyalahedrins. Curr Opin Cell Biol2:839-844, 1990 22. Turley EA: Hyaluronan binding proteins and receptors. Adv Drug Delivery 7:257-264, 1991 23. Oksala 0, Salo T, Tammi R, et al: Expression of proteoglycans and hyaluronan during wound healing. J Histochem Cytochem 43:125135,1995 24. Adzick NS, Longaker MT: Characteristics of fetal repair, in Adzick NS, Longaker MT (eds): Fetal Wound Healing. New York, NY, Elsevier, 1992, pp 53-70
Induces By Joseph
A. locono,
Scarless Repair H. Paul
Ehrlich,
Hershey,
Kerry
Methods: Time-dated pregnant CD-l mice (term, 20 days) were killed on gestational day 18 and fetuses were harvested via laparotomy. A through and through stab wound was made in each forelimb with a l-mm microscapel, and the wound was closed with a single IO-O nylon suture. The forelimbs were amputated at the level of the shoulder and placed in organ culture. Daily medium changes with 1 mL of BGJb (devoid of serum) were made. Half the cultures received 10 uL of HA (4 mg/mL) directly to the wound site with each medium change. The other half of the cultures received 10 uL of phosphate-buffered saline (PBS-control). At day 7, the limbs were harvested, fixed in methyl Carnoys solution, paraffin embedded, and 5-pm serial sections cut. The sections were stained with H&E or Sirius red/fast green. The sections were viewed in a blinded fashion by two observers. Suture defined the wound site, and the sections were graded for healing by scarring.
CAR FORMATION terminates a cascade of events that is catalyzed by an inflammatory response that includes the release of cytokines, hormones, and growth factors. Platelets, neutrophils, and macrophages release these factors within the early wound milieu. The degree to which each component is required in the tissue repair process is unclear. Moreover, alterations in these elements are assumed to play a role in scarless fetal repair. Repair in organ cultured fetal forelimb explants limits the environment to local fibroblasts and their surrounding extracellular matrix. The model facilitates the study of From the Section of Pediatric Surgery, Department of Surgery, Milton S. Hershey h4edical Center; Hershey, PA. This work was supported through grants GM 41343 (TMK) and GM Fl7566 (JAI) from the National Institutes of Health. Address reprint requests to Joseph A. Iocono, MD, Section of Pediatric Surgery, Department of Surgery, Milton S. Hershey Medical Centet; The Penn State G&singer Health System, 500 University Di; Hershey, PA 17033-0850. Copyright o 1998 by W.B. Saunders Company 0022.3468/98/3304-0004$03.00/O
564
A. Keefer,
and
Thomas
Organ M.
Culture
Krummel
Pennsylvania
Background/Purpose:Wounded fetal mouse limbs harvested from two distinct time points in gestation heal differently in organ culture. The healing of a gestational day 14 limb is by scarless repair, whereas gestational day 18 (gd 18) limbs heal by scarring. The persistence of elevated levels of hyaluronic acid (HA) is a major difference in the extracellular matrix of scarless repair. The purpose of this study was to demonstrate that chronic additions of HA to incisional wounds of gd 18 limbs induces scarless repair.
S
in Mouse Limb
Results: Minimal limb growth occurred in both control and HA-treated limbs. Grossly, both control and treated limbs healed incisional wounds by 7 days in culture. Limbs from both treatment and control groups showed viability by microscopic analysis. The limbs treated with HA had no appreciable scar morphologically in sections in which epithelial dimpling and suture were evident. The orientation of the collagen fiber bundles in the control wounds were in parallel arrays perpendicular to the incision. The orientation of the collagen fiber bundles in the HA-treated limbs had a basket weave pattern that was indistinguishable from unwounded dermis. The direct repeated additions of HA to healing organ cultured limb explants of gestational day 18 fetal mice promoted scarless repair. Conc/usions:This result demonstrates that chronic elevation of HA in the microenvironment of a wound affects healing by promoting the deposition of a more dermal-like connective tissue matrix in the wound site. The maintenance of elevated levels of HA could have utility in the clinical setting to improve the organization of connective tissue, leading to the reduction of scar complications. J Pediatr Surg 33:564-567. Copyright o 1998 by W.B. Saunders Company. INDEX culture,
WORDS: scarless
Hyaluronic healing.
acid,
fetal
tissue
repair,
organ
single factors to the wound site and eliminates contributions from the circulation. The isolated, unperfused, fetal mouse limb demonstrates that skin wound repair proceeds in the absence of a circulation. Mouse forelimb organ cultures of gestational day 14 (term, 20 days) heal suture-closed incisional wounds without scar, whereas fetal mouse forelimbs isolated from gestational day 18 healed identical wounds with scar. Fetal mouse forelimb explants remained viable, grew,r and differentiated2 in organ culture. Moreover, the system was shown to be sensitive to added factors such as transforming growth factor beta (TGF-P), in which it promoted repair by scarring in the 14-day limb.3 The effect of daily addition of hyaluronic acid (HA) to healing l&day fetal mouse limbs in organ culture was studied. MATERIALS
AND
METHODS
Mice Time-dated pregnant CD-1 mice (Charles River Laboratories, Wilmington, MA) were individually caged and fed a standard rodent chow and water ad libitum. On gestation day 18 (term, 20). the pregnant
~ourna/ofPediatric
Surgery,
Vol33,
No 4 (April),
1998: pp 564-567
HYALURONAN
INDUCES
SCARLESS
REPAIR
females were killed by cervical dislocation, then hysterectomies performed. The entire uterus with fetuses was placed in ice cold phosphate-buffered saline (PBS). Each fetus was individually removed from its amniotic sac and placed in a petri dish under a dissecting microscope. A linear, l-mm full-thickness incision wound was made in each forelimb. The wound was closed with a single 10-O nylon suture at the midpoint of the incision. Each limb was then amputated at the level of the shoulder and placed in an organ culture dish. The experiment was run in triplicate (n = 12 for each treatment group).
Organ Culture System Organ culture dishes (#3037, Falcon Plastics, Lincoln, NJ), 60 mm X 15 mm designed with a central well and surrounding moat were placed into standard lo-cm petri dishes. The outer moat was filled with sterile water to maintain a humid environment. Stainless steel, wire mesh triangles (Industrial Wire Products, Pasadena, CA) were sterilized, then placed over the central well. A permeable 0.2~pm filter was then placed on top of the grid. BGJ-b medium (Gibco, Gaithersburg, MD), a chemically defined culture medium was supplemented with penicillin G, 1000 U/mL; streptomycin, 0.1 mg/mL; amphotericin B, 0.25 mg/mL; and ascorbate, 1 mg/mL @H adjusted to 7.4 to 7.5 with 1.0 mmol/L NaOH). The central well was filled with 1 mL of the medium, which was even with the surface of the wire mesh triangle. Half the cultures (n = 12) received 10 pL of PBS (control). whereas the other half (n = 12) received 10 pL HA (Sigma, St Louis, MO) to a final concentration of 0.4 mg/mL. The limbs were placed on the membrane, which was placed on the stainless steel mesh, the dish covered and incubated at 37°C in humidified 95% air with 5% CO2 for 7 days. The medium was changed daily.
Documentation of Gross and Histological Changes Gross morphological changes in the limbs were documented by photography on day 7 before harvesting. Histological examination was performed at 7 days by fixing the limbs in methyl Camoys fixative (10% acetic acid, 60% methanol, and 30% chloroform, v/v) for 24 hours, then replacing it with absolute ethanol and stored at -20°C until processing. Sections were embedded in pat&in, 5-pm serial sections were cut through the entire wound and then stained with either H&E or Sirius red with Fast green. Sections were viewed with both bright field and polarized light, and photographs were taken with Kodak Ektachrome 200 ASA slide film. Viability of each limb was confirmed with H&E-stained sections. Collagen bundles were easily visualized by their birefringence pattern exhibited under polarized light. Limbs were scored by two independent observers for wound closure and presence of scarring. Wound was defined in sections with visible suture and epidermal dimpling present. Scarring was defined by the presence of thick parallel collagen bundles beneath the epidermis under polarized light. Absence of scarring was defined by the presence of normal dermal architecture in the wound area and absence of thick parallel bundles of collagen.
565
were reviewed per limb. As expected, there was no evidence of inflammatory response in these organ cultured limbs. In the control group, all 12 limbs showed increased collagen deposition in the healed wound, consistent with tissue repair with scar. When examined under polarized light, the collagen fiber bundles were thicker than surrounding unwounded dermis and in parallel arrays (Fig 1). However, examination of the healed wound of the 12 limbs treated with HA uniformly showed normal dermal and epidermal architecture. When viewed with polarized light, the collagen fiber bundles deposited were in a basket weave pattern, and there was less collagen deposited compared with controls (Fig 2). Little difference in collagen fiber pattern was evident between wound and surrounding dermis. The chronic additions of HA to gd 18 incisional wounds in fetal limbs promoted scarless repair. DISCUSSION
The organ culture explant system facilitates the study of individual variables in the wound healing process within a complete intact organ. The fetal limb contains an organized dermis and epidermis as well as developing muscle and bone. Individual factors can be evaluated with this system. The fetal limb explant is dependent only on bathing culture medium for its nutrients. Several influences, extrinsic to the local wound environment, possibly play a role in regulating tissue repair. The degree of maturation in the fetal immune system and its inability to mount an inflammatory response to injury has been investigated.4 Further, the topical effects of different
RESULTS
In both the PBS-control and HA groups, wound closure was evident in all specimens by gross and histological inspection. Limb length did not appreciably increase, but further differentiation of digits and claw growth were evident (consistent with previous results). Histological differences in the character of the tissue repair between the control and HA-treated groups were found. There were 12 wounded limbs reviewed from each group. An average of 15 slides (20 sections per slide)
Fig 1. A representative cross section of a wounded forelimb from the control group stained with Sirius red as viewed with polarized light. The wound is shown between the arrows. Note the collagen fiber bundle architecture with thick parallel arrays (scar) beneath a regenerated epidermal layer. (Original magnification x400.)
566
Fig 2. A representative cross section of a wounded forelimb from the HA treatment group, stained with Sirius red as viewed with polarized light. The wound is shown between the arrows. The collagen fiber bundle architecture is the same as unwounded dermis. There is no appreciable scar under the regenerated epidermis. (Original magnification x400.)
growth factors found in the amniotic fluid are also under close study.5 They may change as fetal development continues until their increased concentrations may promote scar formation. The organ culture system has the ability to investigate any of these factors alone or in combination. Isolated limb cultures grown in a serumfree medium succeed in eliminating extrinsic factors of tissue repair. It is an unperfused system in which circulating cellular elements are absent. The culture media contains only defined materials where serum and amniotic fluid are excluded. HA is a negatively charged glycosaminoglycan composed of repeated disaccharides of D-glucuronic acid and N-acetylglucosamine.6 HA is the major glycosaminoglycan of the fetal extracellular matrix.7 HA has been shown to influence a number of cellular processes, which include tumor metastasis4 and inflammation8 HA also has been shown to enhance the healing of cornea1 wounds.9J0 Abantangelo et al” demonstrated that, in diabetic rats, HA enrichment of open wounds promoted quicker healing. Alexander and Donoff reported that HA initially accumulated in adult rabbit open wounds but was reduced by the appearance of hyaluronidase (HAdase).12 In contrast, the induction of HAdase was not seen in fetal wounds that healed without scar.13 Scar tissue is defined by the presence of thick bundles in a parallel array beneath a regenerated epidermis. These thick bundles seen in scar are the result of the deposition and organization of collagen fibers. Collagen fiber bundle deposition in a
IOCONO
ET AL
basket weave pattern indistinguishable from the pattern of unwounded dermis defines scarless repair. The amount and organization of collagen fiber bundles can be studied using Sirius red stained serial sections of wounds under polarized light. This technique demonstrates the density of collagen bundles by their intensity and birefringence pattern. As an example, the collagen fibers of normal dermis exhibit a basket weave birefringence pattern, whereas scar collagen fibers are thicker and are parallel to the epidermal layer. Grading of serial sections for collagen intensity and organization of the collagen bundles has been a useful tool for semiquantitative assessment of scarring in 1%day fetal mouse limbs.14 This technique was used to study the effect of chronic HA addition on repair in organ culture. The spatial organization of collagen fiber bundles within connective tissue can be demonstrated by Sirius red staining and polarized light viewing.15 Examination of the color and intensity of the birefringence pattern of collagen fiber bundles provides a visible means for documenting the degree of collagen organization. Immature collagen fiber bundles show fine fibrils having a green birefringence. Advanced, more organized collagen fiber bundles are thicker with a yellow-orange birefringence in a basket weave pattern. Young granulation tissue shows fine collagen fiber bundles in parallel arrays with green birefringence, and more mature granulation tissue or scar shows thick collagen fiber bundles arranged in parallel arrays with yellow-orange birefringence.16 There has been some debate as to whether the maintenance of high levels of HA in wounds that heal scarlessly is a causative factor or epiphenomenon of scarless fetal repair.17 This study shows that HA by itself evokes scarless repair. The fetal extracellular matrix must facilitate the growth, differentiation, and organogenesis of the developing fetus. Fetal matrix is rich in HA.‘* As fetal growth and differentiation nears completion at the end of gestation, HA levels decline.r3 Postnatally, during the normal repair mechanism, HA levels are initially elevated. However, in the adult wound, HA is degraded, whereas in fetal wounds, HA remains elevated.19 Midgestation fetal rabbit wounds, which repair scarlessly, can be induced to heal with scar by treating them with Hyaluronidase (HAdase), which specifically degrades HA.20 The mechanisms whereby HAdase eliminates scarless repair remains unclear. Reducing the levels of HA may affect cell signaling through modulation of the cell surface receptors.21-23 The mechanism for regenerative repair in the 14-day fetal limb explant and the adultlike repair in l&day limb explants is unclear.’ It is proposed that differences in GAG content of the extracellular matrix of the fetal dermis is responsible for these differences in the repair
HYALURONAN
INDUCES
SCARLESS
567
REPAIR
process. The fetal mouse limb organ culture system agrees with the in vivo system of fetal sheep and rabbitsz4 Scarless fetal tissue repair occurs in the intact fetus during early and midgestation (14-day mouse fetal limb). Later in gestation scarless repair is replaced with adult scar repair (1 g-day fetal limb). However, because of its complexity, the in vivo system is difficult to assess mechanistically. The well-defined fetal mouse limb explant organ culture system allows the reduction in the complexity of the system and offers a simpler system for
the study of mechanisms responsible for scarless fetal repair. Maintaining elevated levels of HA in adult wounds may decrease or eliminate scarring. The elucidation of the exact cellular mechanisms whereby HA exerts its effects may be helpful in designing therapeutic interventions to control scarring. The ability to maintain a sustained level of HA for a defined period of time within a wound site may reduce the complications of scarring in the future.
REFERENCES 1. Bleacher JC, Adolph VR, Dillon PW, et al: Isolated fetal mouse limbs: Gestational effects on tissue repair in an unperfused System. J Pediatr Surg 28(1):1312-1315, 1993 2. Chopra V, Blewett CJ, Ehrlich HP, et al: The transition from fetal to adult repair occurring in forelimbs maintained in organ culture. Wound Rep Regen 5:47-51,1997 3. Meisler N, Keefer KA, Ehrlich HP, et al: Dexamethasone abrogates the fibrogenic effect of transforming growth factor-in rat granuloma and granulation tissue fibroblasts. J Invest Derm 108:285-289, 1997 4. Knudson CB, Knudson W: Hyaluronan-binding proteins in development, tissue homeostasis, and disease. FASEBJ 7:1233-1241, 1993 5. Puolakkainen PA. Twardzik DR, Ranchalis JE, et al: The enhancement in wound healing by transforming growth factor-beta 1 TGF-beta 1) depends on the topical delivery system. J Surg Res 58:321-329, 1995 6. Iozzo RV: Proteoglycans: Structure, function and role in neoplasia. Lab Invest 53:373-396, 1985 7. DePalma RL, Krummel TM, Durham LA, et al: Characterization and quantification of wound matrix in the fetal rabbit. Matrix 9:224231,1989 8. Weigel PH, Fuller GM, Le Boeuf RD: A model for the role of Hyaluronic acid and fibrin in the early events during the inflammatory response and wound healing. JTheor Biol119:219-234,1986 9. Nakamura M, Nishida T, Hikida M, et al: Combined effects of hyaluronan and fibronectin on cornea1 wound closure of rabbit in vivo. Curr Eye Res 13:385-388,1994 10. Nishida T, Nakamura M, Mishima H, et al: Hyaluronan stimulates cornea1 epithelial migration Exp Eye Res 53:753-758, 1991 11. Abatangelo G, Martelli M, Vecchis P: Healing of hyaluronic acid enriched wounds: Histological observations. J Surg Res 35:410-416, 1983 12. Alexander SA, Donoff B: The glycosaminoglycans of open wounds. J Surg Res 29:422-426, 1980
13. Krummel TM, Nelson JM, Dieglemann RF, et al: Fetal response to injury in the rabbit. J Pediatr Surg 22:640-644, 1987 14. Cooney RN, Iocono JA, Maish G, et al: Tumor necrosis factor mediates impaired wound healing in chronic abdominal sepsis. J Trauma 42:415-420, 1997 15. Constantine FS, Mowry RW: The selective staining of human dermal collagen: The use of picrosirus red with polarization microscopy. J Invest Derm 50:419-423,1968 16. Ehrlich HP, Desmouliere A, Diegelmann RF: Morphological and immunochemical differences between keloid and hypertrophic scar. Am JPathol 145:105-113, 1994 17. Ferguson MW, Whitby DJ, Shah M, et al: Scar formation: The spectral nature of fetal and adult wound repair. Plast Reconstr Surg 97854-860, 1996 18. Depalma RL, Krummel TM, Nelsom et al: Fetal wound matrix is composed of proteoglycan rather than collagen. Surg Forum 38:626628, 1987 19. Estes JM, Adzick NS, Harrison MR, et al: Hyaluronate metabolism undergoes an ontogenic transition during fetal development: Implications for scar-free healing. J Pediatr Surg 28:1227-1231, 1993 20. Mast BA, Haynes JH, Krummel TM, et al: In vivo degradation of fetal wound hyaluronic acid results in increased fibroplasia, collagen deposition and neovascularization. Plast Reconstr Surg. 89:503-509, 1992 21. Toole BP: Hyaluronan and its binding proteins, the hyalahedrins. Curr Opin Cell Biol2:839-844, 1990 22. Turley EA: Hyaluronan binding proteins and receptors. Adv Drug Delivery 7:257-264, 1991 23. Oksala 0, Salo T, Tammi R, et al: Expression of proteoglycans and hyaluronan during wound healing. J Histochem Cytochem 43:125135,1995 24. Adzick NS, Longaker MT: Characteristics of fetal repair, in Adzick NS, Longaker MT (eds): Fetal Wound Healing. New York, NY, Elsevier, 1992, pp 53-70