Exogenous non-crosslinked collagen enhances granulation tissue formation in dermal excision wounds in guinea pigs

i) ;'., ,'7~'~ ~) 1 ~'q

~ ',c.l.,~~,~ ~

~,~~/~

Exogenous Non-Crosstinked Co[[agen Enhances Granulation Tissue Formation in Derma[ Excision Wounds in Guinea Pigs MEIR REDLICH, HELENA COOPERMAN, HELENAYAKOVLEE REGINA FEFERMANand SHMUEL SHOSHAN Connective Tissue Research Laboratory, Department of Oral Biology, Hebrew University-Hadassah Faculty of Dental Medicine, Jerusalem, Israel

Abstract Based on previous observations indicating a role for collagen peptides in eliciting a positive feedback for collagen biosynthesis, this study was initiated to elucidate the effect of noncrosslinked collagen on granulation tissue formation in dermal excision wounds. The wounds were treated with either non-crosslinked or crosslinked native collagen, or left untreated as controls. Granulation tissue was analyzed for collagen type I mRNA, for levels of interstitial collagen and for the number of blood vessels. The results indicated significant increases in procollagen type I mRNA, in interstitial collagen, in the number of blood vessels and in epithelial advance in the non-crosslinked collagen-treated wounds relative to the untreated controls. It is assumed that the presence of non-crosslinked collagen in a healing wound enhances both procollagen type I biosynthesis and the repair process of dermal wounds, due to the more readily released collagen peptides derived from this exogenous collagen dressing. Key words: non-crosslinked collagen, procollagen type I biosynthesis.

Introduction Repair of injured tissue occurs as a sequence of events in which cells with distinct functions are attracted to the wound, proliferate and secrete components of the extracellular matrix (ECM) to restore structure and function. Following an ordered influx of cells into the injured tissue, an adequate granulation tissue is formed. The cells most prevalent in the repair process are platelets, monocytes/macrophages, fibroblasts, endothelial cells and mast cells, which in turn are recruited by a variety of Abbreviations used: ECM, extracellular matrix. Matrix Biology Vol. 17/1998, pp. 667-671 © 1998 by Gustav FischerVerlag

biologically active substances such as growth factors, eicosanoids, and components of the ECM such as collagen and fibronectin (Clark, 1996). The ECM, mostly produced by fibroblasts, provides a subs[rate that binds other matrix components and facilitates cellular migration, cell adhesion and matrix remodeling (Clark, 1996). Chemotactic migration of fibroblasts during granulation tissue formation is affected both by native collagen and by collagen-derived peptides (Postlethwaite et al., 1978; Albini et al., 1985). Furthermore, in vitro studies have shown that pep[ides derived from collagen breakdown influence the biosynthesis of this protein (Herrmann et al., 1992; Katayama et al., 1993). Also~ in vivo

668

M. Redlich et al.

studies showed a beneficial effect of a native collagen dressing on the healing process of dermal injuries (Shoshan and Yaffe, 1988; De Vries et al., 1995). Thus, it was assumed that collagen derived peptides released from the exogenous collagen dressing may enhance wound repair. Consequently, we assumed that noncrosslinked collagen, which is very soluble and readily degraded, ~vould be more effective than the entire molecule in enhancing wound healing ~vhen applied onto dermal excision wounds. The present study was initiated to investigate the effect of exogenous non-crosslinked collagen on collagen synthesis as well as blood vessel formation in and epithelialization of dermal excision wounds.

Materia[s and Methods The study protocols for the animal experiments were approved by the Animal Care and Use Committee of the Hebrew-University of Jerusalem and followed the National Institutes of Health guidelines, the Care and Use of Laboratory Animals. Guinea pig dermal native non-crosslinked collagen was prepared following daily i.p. administration of ~aminopropionitrile (BAPN), (Sigma, St. Louis, MO), 1 mg/gr body weight for 3 weeks, and subsequent extraction with cold 1 N NaCI solution. Crosslinked collagen was prepared from guinea pig skin by extraction with cold 0.5 M acetic acid. The absence or presence of crosslinks in the preparations was ascertained by SDS-PAGE analysis (sodium dodecyl sulfate polyacrylamide gel electrophoresis). Both extracted collagens were further purified according to Gross (1958), lyophilized and stored in vacuo until used. Before application, a 0.3% collagen solution was prepared in phosphate buffer, at an ionic strength of 0.4 and pH of 7.6. Two full thickness dermal excision wounds were made with a 6 mm biopsy punch (Stiefel, UK) on the shaved backs of 60 ether-anesthetized female guinea pigs weighing 250-280 g. The animals were divided into three groups, as follows: (1) in 20 animals, untreated wounds served as controls; (2) wounds of 20 animals were covered with 0.5 ml of native 0.3% acid-soluble highly crosslinked dermal collagen from normal guinea pigs; (3) wounds of 20 animals were covered with 0.5 ml of native 0.3% dermal collagen solution. This was obtained from lathyritic guinea pigs, i.e., without crosslinks. The different collagens were applied onto the wounds once immediately after wounding. All the

wounds were inflicted under aseptic conditions and were sprayed with OpSite dressing (Smith and Nephew, UK). This adhesive film is being used routinely in our in l,ivo wound healing studies where other bandages are not desirable. Food and water were supplied ad lit)itum. Ten animals of each group were sacrificed with an overdose of pentothal 5 days and 15 days after woundrag. The granulation tissue from ten ~vounds of each group was removed with a biopsy punch, thus ensuring standard samples for analysis. They were immediately frozen m liquid air. Total RNA was isolated from the granulation tissue according to the procedure of Chirgwin (1979), following dispersion of the samples by a high speed Polytron homogenizer for 60 sec. The detailed procedures for Northern blot analyses and Slot blot hybridization and quantification of procollagen type I mRNA were carried out as previously reported (Redlich et al., 1994). Granulation tissue from five wounds of each group was excised, weighed (wet weight) and acid hydrolyzed for hydroxyproline determination as described by Stegemann and Stalder (1967). The results were expressed as tag per mg tissue. Five wounds ~vith surrounding intact skin from each group were dissected, fixed in 4% phosphate-buffered neutral formalin, dehydrated, embedded in formalin and processed for histological and immunohistological analyses. The sections (6 him) were stained with hematoxylin and eosin (H & E) and picrosirius red (PSR). The effect of the treatment ~vas assessed by measuring the epithelial advancement over the excised wound, using a microgrid, and subsequent calculation of the percent of initial ~w~und area. In this study, no attempts were made to measure xvound contraction. The blood vessels in the granulation tissue were counted following incubation of the sections with biotinylated lectin from Bandeirae simplicifoli solution (Sigma, St. Louis, MO), and subsequent treatment with avidin-biotinylated peroxidase complex using a Vectastain ABC Elite kit (Vector laboratories, Burlingame, CA). The results were expressed as number of blood vessels per unit wound area. The statistical significance of the differences between the groups was assessed by the two-sample Student's test for differences in means.

Resu[ts A significant increase in procollagen 0~1(I) mRNA levels following treatment with either non-crosslmked or

Exogenous N o n - C r o s s l i n k e d Collagen Dressing crosslinked collagen, as c o m p a r e d with untreated controis, was noticed after 5 days, 3.67 _+ 1.43 DU in the non-crosslinked collagen treated w o u n d s and 2.73 _+ 0.2 DU in the crosslinked collagen treated w o u n d s , v s 1.06 _+ 0.45 DU in the untreated controls (p < 0.05), (Table IA). The increase in the non-crosslinked collagen treated w o u n d s , although statistically not significant, ~ a s nevertheless higher than in the crosslinked collagen treated w o u n d s . After 15 days, there was a significant increase in the expression of procollagen 0tl(I) m R N A levels in the untreated w o u n d s but not in the treated w o u n d s , relative to that observed after 5 days (Table IA). H y d r o x y p r o l i n e was increased in both non-crosslinked and crosslinked collagen-treated w o u n d s as compared to the untreated w o u n d s 5 days after w o u n d i n g (13.37 _+ 7.96 and 9.94 -+ 4.49, respectively, v s 2.28 _+

Table 1. A. Expression of steady-state levels of procollagen type I mRNA 5 days and 15 days after wounding.

Untreated controls Non-crosslinked collagen (rosslinked collagen

5 days

15 days

1.06 _+0.45 3.67 * 1.43" 2.73 _+0.2*

5.31 _+ 1.74 ...... 4.27 .+ 0.98 3.21 _+0.67

Each value represents Mean +_ S.D. (in densitometric units, DU.) * p < 0.05 relative to untreated controls. ': * p < 0.05 relative to 5 days.

669

1.52; p < 0.05). After 15 days, the a m o u n t of h y d r o x yproline significantly increased in all the groups as comp a r e d to 5 days (Table IB). Histological e x a m i n a t i o n revealed a conspicuous enh a n c e m e n t in w o u n d closure in the non-crosslinked collagen-treated w o u n d s relative to the untreated controls after 5 days: 87 _+ 3 . 9 % v s 27.5 _+ 1.6%, p < 0.05. W o u n d closure in the non-crosslinked collagen-treated w o u n d s was also significantly advanced relative to that of the crosslinked collagen treated w o u n d s (Fig. 1A). The n u m b e r of b l o o d vessels in g r a n u l a t i o n tissues of the non-crosslinked treated w o u n d s was significantly higher than in the untreated w o u n d s after 5 days, but it was lower after 15 days (Fig. 1B and Table IC). Discussion The present study showed that non-crosslinked collagen had a conspicuous effect both on e p i d e r m a l advance in dermal excision w o u n d s and on the n u m b e r of b l o o d vessels, as well as on the remodeling stage as shown by procollagen type I synthesis and deposition in the granulation tissue 5 days after w o u n d i n g . It is assumed that this effect of the non-crosslinked collagen was b r o u g h t a b o u t by readily released collagen peptides which have an immediate and rapid effect on fibroblasts and extracellular proteins, as indicated in earlier studies (Albini et

120

B. Hydroxyproline (lag/mg tissue) in granulation tissues 5 days and 15 days after wounding. 5 days Untreated controls Non-crosslinked collagen Crosslinked collagen

15 days

2.28 _+ 1.52 13.37 _+7.96* 9.94 _+6.49*

0.05 relative to untreated controls. 5 days. * p <

* * p <

22.68 + 9.83 ...... 39.18 .+ 9.7* * 34.96 _+9.11"*

100

~// /// O0

~)

o

Z

80

,*x*C ,*.*.,

[]

5/~ :5::::

[]

x-linked

[]

non-x-lnkd

I I i I I I I I I ~ ~

~ q

~J~

60 •

,'.,',,

0.05 relative to

~ i / i / / I I / /

/// //J ~/J

40 e,~

///

C. Number of blood vessels in granulation tissues per unit area, 5 days and 15 days after wounding.

f j / ~

20

~

,'~*,, * 1 , . ~ ~~ , ~ , , ~ ~ , , ~

unlr.¢trl

, ~ ~ , , ~ ~ , , ~ ~ , . ~ I l l . , ~ t Ij . ~ # 1 , . ~ . * ~* ,

*~* . ~

¢¢, . ~

p # , . ~ ~ , . ~ . ~ ~ , . ~ . ~ . ~

5 days

15 days

0.08 _+0.02 1.6 _+0.10 *°:1.3 .+ 0.09*

0.9 .+ 0.04 0.5 _+0.03** 1.0 _+0.06

15 days

5 days

Untreated controls Non-crosslinked collagen Crosslinked collagen

Five wounds per group per time point. * p < 0.05 relative to untreated controls. ** p < 0.005 relative t o untreated controls.

days

Figure IA. Effect of non-crosslinked collagen on wound closure expressed as percent of initial wound area. * = p < 0.01 relative to untreated control ** = p < 0.005 relative to untreated control

670

M. Redlich et al.

~

~ ,.:

...........

~

~

,

::: 2

,

,,

.~

~,~

Figure lB. (A) Photomicrograph of untreated control to sho,v blood vessels in the granulation tissue 5 days after wounding. Avidin-Biotinylated peroxidase stain (for details see text). Original magnification 320x. (B) Photomicrograph of non-crosslinked collagen-treated wound 5 days after wounding to show abundance of blood vessels. Original magnification 400x. (C) Photomicrograph of non-crosslinked collagen-treated wound 15 days after wounding to shoxv a conspicuous reduction in the number of blood vessels in the granulation tissue. Original magnification 400x.

al., 1985; Shoshan and Yaffe, 1988; De Vries 1995). Thus, it is assumed that the healing wound through the various healing stages (inflammation, ithelialization, matrix formation and remodeling)

et al., passes re-epfaster

following application of non-crosslinked collagen than in untreated wounds or in wounds treated with crosslinked collagen. The enhancing effect of the noncrosslinked collagen on the healing process was also re-

Exogenous Non-Crosslinked Collagen Dressing flected in blood vessel count. The observed highest number of blood vessels in the granulation tissue of noncrosslinked collagen-treated wounds after 5 days and its dramatic decrease after 15 days indicate a significantly earlier advanced remodeling stage as c o m p a r e d to both crosslinked collagen treated and untreated controls. The increase in the expression of procollagen c~l(I) m R N A and in intersitial collagen at day 15 relative to day 5 in the untreated controls agrees with an earlier description of a time-dependent increase in collagen biosynthesis during the first 2 weeks after injury (Oono et al., 1993). The present study has shown that exogenous non-crosslinked collagen enhances this process. The observed increase in interstitial collagen in both non-crosslinked and cross-linked collagen was apparently not due to the applied collagen, which is removed from the w o u n d after 11-12 days, as shown in earlier studies (Shoshan and Finkelstein, 1970). We conclude that exogenous non-crosslinked collagen affects the healing process by advancing the proliferative stage and shifting the remodeling stage of a healing wound to an earlier time point. However, one has to look at other E C M components which are known to influence the healing process and with which the exogenous collagen may be interacting, such as fibrin, fibronectin, matrix metalloproteinases (MMPs) and elastin, as well as inflammatory cytokines, etc. Indeed, investigations of these interactions are currently in progress.

References AIbini, A. and Adelmann-Grill, B.C.: Collagenolytic cleavage products of collagen type I as chemoattractants for human dermal fibroblasts. Eur. J. Cell Biol. 36: 104-107, 1985. Chirgwin, J.M., Przybyla, A.E., MacDonland, R.J. and Rutter W.J.: Isolation of biologically active ribonucleic acid from sources enriched in ribonucleases. Biochemistry 18: 5294-5299, 1979.

671

Clark, R.A.F.: Wound repair. In: The Molecular and Cellular Biology of Wound Repair, ed. by Clark.,R.A.E, Plenum Press, New York and London, 1996, pp. 3-35. De Vries, H.J.C., Zeegelaar, J.E., Middelkoop, E., Gijsbers, G., Van Marie, J., Wildevuur, C.H.R. and Westerhof, W.: Reduced wound contraction and scar formation in punch biopsy wounds. Native collagen dermal substitutes. A clinical study. Brit. J. Dermatol. 132: 690-697, 1995. Gross J.: Studies on the formation of collagen. I. Properties and fractionation of neutral salt extract of normal guinea pig connective tissue. J. Exper. Med. 107: 247-263, 1958. Herrmann, K., Munzberger, Ch., Krieg, T. and Haustein, U.E: The 23 kDa collagen type I split product modulates the proliferation, matrix production synthesis and integrin expression of human dermal fibroblats. Eur. Soc. I)ermatol. Res., London; 1992. Katayama, K., Armendariz-Borunda, J., Raghow, R., Kang, A.H. and Seyer, J.M.: A pentapeptide from type I procollagen promotes extracellular matrix production..l. Biol. Chem. 268: 9941-9944,1993. Oono , T., Specks, U., Maje~vski, S., Hunzelmann, N., Timpl, R. and Krieg, T.: Expression of type IV collagen mRNA during wound healing. J. Invest. Dermatol. 100: 329-334, 1993. Postlethwaite, A.E., Seyer, J.M. and Kang, A.H.: Chemotactic attraction of human fibroblasts to type 1, II and 111 collagen and collagen-derived peptides. Proc. Natl. Acad. Sci. USA 75: 871-875, 1978. Redlich, M., Peleg, I., Cooperman, H. and Shoshan, S.: Topological differences in the expression of collagen type I and collagen type III mRNAs in the rat gingiva. J. Periodontol. 5: 776-780, 1994. Shoshan, S. and Finkelstein, S.: Acceleration of wound healing induced by enriched collagen solutions. J. Surg. Res. 10: 485-491,1970. Shoshan, S. and Yaffe, A.: Use of collagen solutions in surgery and adjuvant to prosthetic implants. In: Collagen, ed. by Nimni M.E., vol. II1, CRC Press, Boca Raton, Florida, 1988, pp. 209-221. Stegemann, H. and Stalder, K.: Determination of hydroxyproline. Clin. Chim. Acta 18: 267-273, 1967. Dr. Shmuel Shoshan, Connective Tissue Research Laboratory, Department of Oral Biology, Hebrew University-Hadassah Faculty of Dental Medicine, P.O. Box 12270, Jerusalem 91220, Israel. Received June 2, 1998; accepted October 6, 1998