Fetal Rat Amniotic Fluid: Transforming Growth Factor β and Fibroblast Collagen Lattice Contraction

Journal of Surgical Research 100, 205–210 (2001) doi:10.1006/jsre.2001.6243, available online at http://www.idealibrary.com on

Fetal Rat Amniotic Fluid: Transforming Growth Factor ␤ and Fibroblast Collagen Lattice Contraction Howard Levinson, M.D.,* ,1 Ziv Peled, M.D.,* Wei Liu, M.D., Ph.D.,* Michael T. Longaker, M.D.,* Gretchen M. Allison, B.S.,† and H. Paul Ehrlich, Ph.D.† *Department of Surgery, Laboratory of Developmental Biology and Repair, New York University Medical Center, New York, New York; and †Division of Plastic Surgery, Penn State University College of Medicine, Milton S. Hershey Medical Center, Hershey, Pennsylvania 17033 Submitted for publication January 8, 2001; published online August 21, 2001

14 AF contributes to enhanced FPCL contraction. Background. In several mammalian animal models, early-gestational-age fetal wounds heal without scar, but wounds of late gestational age heal with scar. This change in wound healing phenotype can be a result of both intrinsic (i.e., cellular characteristics) and extrinsic (i.e., environmental) factors. Our question was: Does amniotic fluid (AF) influence the change from scarless to scar-forming repair in the rat? Methods. Rat AF was investigated for its modulation of fibroblast-populated collagen lattice (FPCL) contraction and morphological changes of adult fibroblasts. AF was also assayed for transforming growth factor ␤ (TGF-␤) levels. Adult rat dermal fibroblasts in monolayer and incorporated into FPCLs were incubated with AF additions from gestational age 14, 16, 18, and 21 days at 10% (v/v). Results. Day 14 AF significantly stimulated FPCL contraction, but AF of 16, 18, and 21 days inhibited FPCL contraction. Fluorescence histology identified microtubules and microfilaments in AF treated adult rat dermal fibroblasts. The staining pattern of microtubules in Day 14 AF-treated fibroblasts showed denser structures at the cell center. Cells incubated with Day 16 or 18 AF showed fine peripheral microtubules. A mink lung epithelial cell bioassay was used to analyze concentrations of TGF-␤ in AF. TGF-␤ levels were greatly elevated in Day 14 AF, but were relatively low in Day 16, 18 and 21 AF. The inhibitor of FPCL contraction from AF of Days 16, 18, and 21 was not identified. Conclusion. It is proposed that the robust expression of TGF-␤ or cytoskeletal changes induced by Day 1

To whom correspondence should be addressed at Division of Plastic Surgery, H071, The Milton S. Hershey Medical Center, 500 University Drive, Hershey, PA 17033-0850.

© 2001 Academic Press

Key Words: amniotic fluid; collagen lattice contraction; transforming growth factor ␤; fetal wound healing; microtubules; microfilaments. INTRODUCTION

In the fetal rat, the transition from the scarless to the scar-forming phenotype occurs after Day 16 of gestation [1]. Both the intrinsic properties of the fetal cells and the extrinsic environment may contribute to this phenomenon [2– 4]. Intrinsically, a diminished inflammatory response, decreased angiogenesis, and minimal deposition of newly synthesized connective tissue are characteristic of fetal wound healing [1]. Extrinsically, the fetus is continually bathed in amniotic fluid (AF), which contains high levels of hyaluronic acid, fibronectin, and growth factors. AF is also known to influence the healing process in a species-dependent manner [5–10]. When exposed to AF, neither fetal rat wounds nor fetal rabbit wounds contract in vivo [4, 7]. In contrast, fetal sheep wounds do contract in vivo [8]. The in vitro fibroblast-populated collagen lattice (FPCL) contraction model attempts to characterize factors contained in AF that contribute to the in vivo wound contraction phenomenon. FPCLs consist of cultured fibroblasts suspended in a collagen matrix. Fibroblast-directed reorganization of collagen fibers causes a reduction in FPCL size, referred to as FPCL contraction [11]. It is proposed that the in vitro FPCL contraction model simulates in vivo wound contraction. By including soluble additions in the casting of FPCLs, one can document the stimulatory or inhibitory properties of these substances on FPCL contraction. Hu-

205

0022-4804/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.

206

JOURNAL OF SURGICAL RESEARCH: VOL. 100, NO. 2, OCTOBER 2001

man and rabbit amniotic fluid inhibit FPCL contraction, while sheep AF promotes FPCL contraction [4, 8, 9, 12]. Temporal based studies illustrate that human AF has a gestational age-dependent effect on FPCL contraction [9]. There appears to be more than one factor that effects gestational age-dependent FPCL contraction [13]. Differential cytokine expression may play a role. More specifically, transforming growth factor ␤ (TGF-␤), a cytokine known to promote FPCL contraction, may contribute to AF modulation of FPCL contraction [14 –17]. By immunohistology, Whitby and Ferguson report that TGF-␤ is present in adult wounds but absent in fetal wounds [18]. TGF-␤ is an important constituent of adult wound healing and its absence in the fetal wound may contribute to scarless wound healing. Moreover, the exogenous application of TGF-␤ to fetal wounds promotes healing by scar formation [19]. Therefore, TGF-␤ levels appear to influence the fetal wound healing phenotype [19]. To examine whether rat AF contains TGF-␤ late in gestation, a TGF-␤ bioassay was done on AF obtained from rats of different gestational ages. Also, AF was included in the casting of FPCLs and its effect on FPCL contraction was measured. The primary question was: Does late-gestational-age AF contains greater amounts of TGF-␤ than early-gestational-age AF, coincidental with the time at which fetal wounds heal with scar formation? METHODS

Amniotic Fluid Harvest All animal work was approved by the institution’s guide for the care and use of laboratory animals. A total of six rats at each gestational age were used. Rats of gestational ages 14, 16, 18, and 21 days were anesthetized and euthanized with ketamine, rompun, and acepromazine. The fetal sacs were removed from the uterus and a sterile 21-gauge needle was inserted to aspirate AF under sterile conditions. AF was then pooled from all fetal sacs in each gestational age group. On isolation, the AF was frozen in liquid nitrogen and stored at ⫺80°C until studied.

Casting FPCLs and Measurement of Lattice Contraction Medium. Dulbecco’s modification of Eagle’s medium (DMEM) and fetal bovine serum (FBS) were purchased from Life Technologies (Rockville, MD). Complete DMEM contained 10% FBS and 10 ␮g/ml gentamicin. Cell line. Normal rat dermal fibroblasts were derived from the outgrowth from adult Sprague–Dawley rat dermal explants. Fibroblasts were maintained in complete DMEM and studied between their third and fifth passages. FPCLs were manufactured with 100,000 rat dermal fibroblasts suspended in 1.5 ml of complete DMEM containing 200 ␮l of AF or saline. The cell suspension was combined with 0.5 ml of collagen solution (2.5 mg in 1 mM HCl), making a total volume of 2.0 ml. The solution was rapidly mixed, then 0.5-ml aliquots were delivered to 16-mm-diameter wells of 24-well tissue culture plates. Tissue culture plates were placed in a 37°C incubator containing a water saturated atmosphere with 95% air and 5% CO 2, where the collagen polymer-

ized in less than 90 s. Thirty minutes after casting, the FPCLs were detached from the wells with a glass rod. The diameter of each FPCL was measured with a stereomicroscope having a ruler in place. The diameter of each FPCL was recorded daily and the area of each FPCL calculated. Student’s paired t test was used to determine statistical significance at P ⱕ 0.05.

Mink Lung Epithelial Cell TGF-␤ Bioassay The method has been previously described [20]. Briefly, mink lung (ML) cells were stably transfected with a plasmid containing the luciferase gene (a gift of Dr. D. B. Rifkin, NYU Medical Center) downstream of a TGF-␤-sensitive portion of the plasminogen activator inhibitor 1 (PAI-1) promoter. TGF-␤ stimulation of ML cells activates the PAI-1 promoter and subsequent synthesis of luciferase. ML transfected cells were suspended at 2 ⫻ 10 4cells/100 ␮l of complete DMEM and plated at 100 ␮l per well in a 96-well plate. ML cells were allowed to attach for 3– 4 h before the medium was removed. FBS-free DMEM supplemented with 0.1% bovine serum albumin, 10,000 units penicillin/ml, and 10,000 ␮g streptomycin/ml was added to control wells. The same supplemented DMEM was used for all serial dilutions of AF. This bioassay measured total TGF-␤ (TGF␤ 1–3) activity in the conditioned media in both latent and active forms [20]. Latent and active TGF-␤ were assessed by analyzing untreated and heat-treated (80°C for 10 min) AF, respectively. TGF-␤ was serially diluted and added in triplicate as a positive control. A negative control was established, where TGF-␤ pan-neutralizing antibody (20 ␮g/ml) was added to a separate group of heat activated AF. The plates were incubated for 16 h at 37°C. The medium was removed from each well. Each well was washed three times with phosphate-buffered saline (PBS). ML cells were lysed and luminescence read by a luminometer. The experiment was done in duplicate on the pooled AF from rats of gestational ages 14, 16, 18, and 21 days. ANOVA was used to determine statistical significance at P ⱕ 0.05.

Fluorescence Histology Adult rat fibroblasts were plated on 22-mm-square coverslips at 5,000 cells/cm 2 in complete DMEM containing AF of different gestational ages at 10% (v/v). After 24 h, incubated cells were fixed in buffered 4% paraformaldehyde in PBS at room temperature for 5 min. The immunofluorescent staining of microtubules used a mouse monoclonal antibody directed against ␣-tubulin (Sigma Chemical Co., St Louis, MO). The cells were rinsed in PBS and permeablized with 0.1% Triton X-100 (Sigma Chemical Co.) for 1 min. After PBS rinses, antibody was added and incubated for 1 h at room temperature. After three washes with PBS, a secondary goat anti-mouse IgG polyclonal antibody conjugated with Oregon Green “488” (Molecular Probes, Inc., Eugene, OR) was included for 1-h. After three washes with PBS, mountant was added and a glass slide placed on top. Other AF-treated fibroblasts were grown on coverslips for 24 h and fixed the same as tubulin-stained cells. To demonstrate cytoplasm microfilament structures, fixed cells were incubated with rhodamine (Rh)– phalloidin following the manufacturer’s instructions (Molecular Probes, Inc.). Fluorescent microtubules and microfilaments were viewed with a Zeiss inverted fluorescence microscope with appropriate fluorescence filters.

RESULTS

FPCL Contraction with Amniotic Fluid The influence of AF from gestational ages 14, 16, 18, and 21 days on FPCL contraction is summarized in Fig. 1. At Day 1, AF from gestational age 14 days significantly enhanced FPCL contraction as compared with PBS-treated controls. In contrast, AF from gesta-

207

LEVINSON ET AL.: RAT AMNIOTIC FLUID

FIG. 1. Collagen lattices populated with adult rat dermal fibroblasts, FPCLs, were exposed to AF from rats of gestational ages 14, 16, 18, and 21 days. Within 24 h, Day 14 AF stimulated FPCL contraction significantly more than controls (P ⱕ 0.05), and Day 16, 18, and 21 AF significantly inhibited FPCL contraction (P ⱕ 0.05). AF stimulation or inhibition FPCL contraction was less effective with 72-h-old FPCL.

tional ages 16, 18, and 21 days significantly inhibited FPCL contraction as compared with PBS-treated controls. These findings were contrary to our hypothesis. Maximal stimulation and maximal inhibition of FPCL contraction occurred within 24 h of exposure to AF. The effects of AF on FPCL contraction diminished over the ensuing 48 –72 h. By Day 3, differences in FPCL contraction between treatment groups were minimal and approached levels equivalent to those of PBS control FPCLs. TGF-␤ Levels in Amniotic Fluid TGF-␤ levels in AF may influence FPCL contraction. AF contains active and inactive (bound-heat activated) forms of TGF-␤. Active baseline TGF-␤ activity was assayed by incubating ML cells in unheated AF. Active and inactive (total) TGF-␤ was assayed by first heating AF before incubation with ML cells. Heat treatment released TGF-␤ from its bound state; hence the levels of luciferase expression reflected total (bound ⫹ unbound) TGF-␤ in the sample. The inclusion of TGF-␤ neutralizing antibody completely blocked AF stimulation of luciferase activity, demonstrating assay specificity for TGF-␤ (data not shown). AF from gestational age 14 days had the highest level of TGF-␤ as compared with AF from gestational ages 16, 18, and 21 days (Fig. 2) (P ⱕ 0.005). The level of TGF-␤ was highest in AF associated with scarless repair and lowest in AF associated with scarring. These findings contradicted our original hypothesis. Microfilaments and Microtubules of Adult Rat Fibroblasts Incubated with AF Adult rat dermal fibroblasts were grown in monolayer for 1 day in the presence of AF of different gestational ages. No differences in microfilaments orga-

nized in stress fibers were obvious between fibroblasts incubated with AF from the three time points (Fig. 3). All treated fibroblasts appeared identical to PBStreated controls. However, the characteristics of microtubules in fibroblasts immunostained with ␣-tubulin antibody were altered in the presence of AF from gestational age 14 days (Fig. 3). The microtubules of adult rat dermal fibroblasts bathed in Day 14 AF were thicker in the perinuclear area of the cell in comparison to peripherally located microtubules in fibroblasts immersed in Day 16 or 18 AF. Cells exposed to Day 16 and 18 AF and PBS had finer microtubule structures. It appears that Day 14 AF affects the organization of microtubules in fibroblasts differently than Day 16 or 18 AF. DISCUSSION

Reported differences between fetal and adult wound healing include: (1) decreased inflammatory response in the fetus, (2) decreased angiogenesis in the fetus, (3) differential extracellular matrix composition, and (4) altered cytokine expression profiles [1]. An intriguing difference between early-gestational-age fetal cutaneous wound repair and adult wound repair is healing without scarring in the fetus. Studies show that both intrinsic (e.g., inflammatory response) and extrinsic (e.g., amniotic fluid) factors may account for these differences in wound healing [1– 4, 21]. Here the focus is on the effects of AF, an extrinsic factor, on the wound contraction phenomenon. Species and age determine whether an open fetal wound will close by wound contraction [8, 9]. Excisional wounds in fetal rabbits and fetal rats do not heal by wound contraction, while open wounds in fetal sheep close by wound contraction [7, 10, 12, 22]. Human AF inhibits FPCL contraction maximally at 21 weeks of gestation [9, 13]. Hence, the transition from fetal to adult-like wound repair may be influenced by developmental changes in AF composition. In this study, the effects of Day 14, 16, 18, and 21 AF on FPCL contraction do not parallel the effects of fetal rat wound contraction in vivo [23]. In vitro Day 14 rat AF stimulates FPCL contraction, while Day 16, 18, and 21 AF inhibits FPCL contraction. Interestingly, FPCL contraction proceeded more slowly in response to Day 16, 18, and 21 AF. Apparently, the AF promoter(s) and AF inhibitor(s) of FPCL contraction were reversible and had a half-life of less than 3 days. In vivo, the Day 14 AF effect is unexpected since excisional wounds made in fetal rats do not contract. The bathing of Day 14 fetal excisional wounds with AF does not promote wound contraction. By an in vitro assay, rat Day 14 AF stimulates contractile activity, but AF derived later in gestation inhibits contractile activity. Rat AF mimics the properties of sheep, rabbit, and human AF both by promoting and inhibiting FPCL contraction, albeit in

208

JOURNAL OF SURGICAL RESEARCH: VOL. 100, NO. 2, OCTOBER 2001

FIG. 2. Luciferase activity of ML cells exposed to AF from rats of gestational ages of 14, 16, 18, and 21 days. (A) Active TGF-␤ (unbound, not heat treated) in AF. More luciferase activity correlates with higher levels of TGF-␤. Serial dilutions were done for experimental values to correlate with positive controls. PBS did not stimulate luciferase expression (data not shown). Day 14 AF stimulated luciferase expression more than Day 16, 18, and 21 AF and control, indicating that more TGF-␤ was active. (B) Total TGF-␤ (bound ⫹ unbound, heat treated). Day 14 AF stimulated luciferase expression more than Day 16, 18, and 21 AF and control, indicating more total TGF-␤.

an age-dependent manner. The contractile-promoting FPCL properties of younger rat AF are similar to those of a 40,000-molecular-weight component identified in sheep AF [10]. Older gestational rat AF, which inhibits FPCL contraction, mirrors a low-molecular-weight protein inhibitor reported in rabbit AF [12]. The hypothesis tested is scarring in late-gestationalage fetal rat cutaneous wounds is due in part to greater amounts of TGF-␤ contained in older gestational AF. Since TGF-␤ is known to enhance FPCL contraction, it was decided to examine TGF-␤ activity in rat AF. Surprisingly, TGF-␤ was elevated (statistically significant) in rat AF of 14 days’ gestation relative to Days 16, 18, and 21. This was paradoxical, since TGF-␤ promotes scarring in adults, yet Day 14 fetal rat wounds heal without scar. It is reported that fetal mouse wounds of 14 days’ gestation do not demonstrate TGF-␤ by immunohistology [18]. TGF-␤ stimulates FPCL contraction influencing the

functioning of microtubules [11, 24]. Microtubule staining reveals that Day 14 AF causes microtubules to form thick bundles in the periphery of treated fibroblasts. AF from days 16 and 18 of gestation inhibits FPCL contraction and promotes fine microtubules. Dynamic microtubules are necessary for optimal FPCL contraction. Therefore, Day 14 AF may enhance FPCL contraction by promoting the dynamic actions of microtubules [24]. TGF-␤ activity is more prevalent in earlygestational-age AF, when the bioassay demonstrated 5-fold more endogenous TGF-␤ activity in AF from Day 14 fetuses relative to AF from Days 16, 18, and 21. Fetal rat skin’s diminished response to extrinsic TGF-␤ contained within AF suggests one possibility that the expression of TGF-␤ receptor II (TGF-␤ RII) may be diminished. Soo et al. have found that TGF-␤ RII expression, which is essential for TGF-␤ signaling, is less at gestational age Day 16 as compared with gestational

LEVINSON ET AL.: RAT AMNIOTIC FLUID

209

FIG. 3. These photographs illustrate fluorescent phalloidin and ␣-tubulin staining in adult rat fibroblasts exposed to amniotic fluid. (A, C, E) ␣-Tubulin staining in fibroblasts exposed to Day 14, 16, and 18 day, respectively. (B, D, F) Phalloidin staining in fibroblasts exposed to Day 14, 16, and 18 AF, respectively. Adult rat fibroblasts exposed to Day 14 AF stain more intensely for centrally located ␣-tubulin (A), whereas fibroblasts exposed to Day 16 and 18 AF demonstrate a more peripheral based staining of ␣-tubulin (C, E). Phalloidin homogenous staining is present in each group of fibroblasts (B, D, F).

age Day 18 [25]. Low levels of TGF-␤ RII in earlygestational-age fetal rat dermal fibroblasts indicate that fetal rat skin has a limited capacity to respond to TGF-␤ stimulation. CONCLUSIONS

Our hypothesis, late gestational age fetal rat AF contains greater amounts of TGF-␤ contributing to scar-forming repair in later-gestational-age fetal rats, is rejected. The transition between scarless and scar repair in the fetus may depend on fetal dermal fibro-

blasts’ ability to respond to TGF-␤, through differential expression of TGF-␤ receptor levels. The mechanisms for generating scarless repair in the fetus remain unclear, yet AF may contribute to the scarless phenotype. Research aimed at manipulating adult wound healing and producing scarless repair by information generated from the study of fetal repair will continue. ACKNOWLEDGMENTS Special thanks to Dr. John Munger and Dr. Dan Rifkin for their support and gift of mink lung cells expressing the TGF-␤ reporter system.

210

JOURNAL OF SURGICAL RESEARCH: VOL. 100, NO. 2, OCTOBER 2001

REFERENCES 1. 2.

3.

4.

5. 6.

7.

8.

9.

10.

11.

12.

13.

Mackool, R. J., Gittes, G. K., and Longaker, M. T. Scarless healing: The fetal wound. Clin. Plast. Surg. 25: 357, 1998. Longaker, M. T., Burd, D. A., Gown, A. M., Yen, T. S., Jennings, R. W., Duncan, B. W., Harrison, M. R., and Adzick, N. S. Midgestational excisional fetal lamb wounds contract in utero. J. Pediatr. Surg. 26: 942, 1991; discussion 26: 947, 1991. Ferguson, M. W., Whitby, D. J., Shah, M., Armstrong, J., Siebert, J. W., and Longaker, M. T. Scar formation: The spectral nature of fetal and adult wound repair. Plast. Reconstr. Surg. 97: 854, 1996. Ledbetter, M. S., Morykwas, M. J., Ditesheim, J. A., Vander Ark, W. D., La Rosee, J. R., and Argenta, L. C. The effects of partial and total amniotic fluid exclusion on excisional fetal rabbit wounds. Ann. Plast. Surg. 27: 139, 1991. Somasundaram, K., and Prathap, K. Intra-uterine healing of skin wounds in rabbit foetuses. J. Pathol. 100: 81, 1970. Sancho, M. A., Julia, V., Albert, A., Diaz, F., and Morales, L. Effect of the environment on fetal skin wound healing. J. Pediatr. Surg. 32: 663, 1997. Belford, D. A. The mechanism of excisional fetal wound repair in vitro is responsive to growth factors. Endocrinology 138: 3987, 1997. Burd, D. A., Longaker, M. T., Rittenberg, T., Adzick, N. S., Harrison, M. R., and Ehrlich, H. P. In vitro foetal wound contraction: The effect of amniotic fluid. Br. J. Plast. Surg. 44: 302, 1991. Wider, T. M., Yager, J. S., Rittenberg, T., Hugo, N. E., and Ehrlich, H. P. The inhibition of fibroblast-populated collagen lattice contraction by human amniotic fluid: A chronologic examination. Plast. Reconstr. Surg. 91: 1287, 1993. Rittenberg, T., Longaker, M. T., Adzick, N. S., and Ehrlich, H. P. Sheep amniotic fluid has a protein factor which stimulates human fibroblast populated collagen lattice contraction. J. Cell. Physiol. 149: 444, 1991. Bell, E., Ivarsson, B., and Merrill, C. Production of a tissue-like structure by contraction of collagen lattices by human fibroblasts of different proliferative potential in vitro. Proc. Natl. Acad. Sci. USA 76: 1274, 1979. Krummel, T. M., Erlich, H. P., Nelson, J. M., Michna, B. A., Thomas, B. L., Haynes J. H., Cohen, I. K., Diegelmann, R. F. In vitro and in vivo studies on the inability of fetal rabbit wounds to contact. Wound Repair Regen. 1: 15, 1993. Yager, J. S., Hugo, N. E., and Ehrlich, H. P. Inhibition of fibroblast-populated collagen lattice contraction by an albuminbound lipid fraction in human amniotic fluid. Plast. Reconstr. Surg. 101: 6, 1998.

14.

Mustoe, T. A., Pierce, G. F., Thomason, A., Gramates, P., Sporn, M. B., and Deuel, T. F. Accelerated healing of incisional wounds in rats induced by transforming growth factor-beta. Science 237: 1333, 1987.

15.

Younai, S., Venters, G., Vu, S., Nichter, L., Nimni, M. E., and Tuan, T. L. Role of growth factors in scar contraction: An in vitro analysis. Ann. Plast. Surg. 36: 495, 1996.

16.

Murata, H., Zhou, L., Ochoa, S., Hasan, A., Badiavas, E., and Falanga, V. TGF-beta3 stimulates and regulates collagen synthesis through TGF-beta1-dependent and independent mechanisms. J. Invest. Dermatol. 108: 258, 1997.

17.

Montesano, R., and Orci, L. Transforming growth factor beta stimulates collagen-matrix contraction by fibroblasts: Implications for wound healing. Proc. Natl. Acad. Sci. USA 85: 4894, 1988.

18.

Whitby, D. J., and Ferguson, M. W. Immunohistochemical localization of growth factors in fetal wound healing. Dev. Biol. 147: 207, 1991.

19.

Bleacher, J. C., Adolph, V. R., Dillon, P. W., and Krummel, T. M. Fetal tissue repair and wound healing. Dermatol. Clin. 11: 677, 1993.

20.

Munger, J. S., Huang, X., Kawakatsu, H., Griffiths, M. J., Dalton, S. L., Wu, J., Pittet, J. F., Kaminski, N., Garat, C., Matthay, M. A., Rifkin, D. B., and Sheppard, D. The integrin alpha v beta 6 binds and activates latent TGF beta 1: A mechanism for regulating pulmonary inflammation and fibrosis. Cell 96: 319, 1999.

21.

Longaker, M. T., Whitby, D. J., Ferguson, M. W., Lorenz, H. P., Harrison, M. R., and Adzick, N. S. Adult skin wounds in the fetal environment heal with scar formation. Ann. Surg. 219: 65, 1994.

22.

Mast, B. A., Diegelmann, R. F., Krummel, T. M., and Cohen, I. K. Scarless wound healing in the mammalian fetus. Surg. Gynecol. Obstet. 174: 441, 1992.

23.

Hallock, G. G., Rice, D. C., Merkel, J. R., and DiPaolo, B. R. Analysis of collagen content in the fetal wound. Ann. Plast. Surg. 21: 310, 1988.

24.

Kolodney, M. S., and Elson, E. L. Contraction due to microtubule disruption is associated with increased phosphorylation of myosin regulatory light chain. Proc. Natl. Acad. Sci. USA 92: 10252, 1995.

25.

Soo, C., Hu, F. Y., Zhang, X., Wang, Y., Beanes, S. R., Lorenz, H. P., Hedrick, M. H., Mackool, R. J., Plaas, A., Kim, S. J., Longaker, M. T., Freymiller, E., and Ting, K. Differential expression of fibromodulin, a transforming growth factor-beta modulator, in fetal skin development and scarless repair. Am. J. Pathol. 157: 423, 2000.