The Cellular Biology of Wound Healing

M.C. REGAN and A. BARBUL

Abstract

Both in vitro and in vivo studies have demonstrated that the presence of both macrophages and T lymphocytes at the wound site is essential in order for the normal healing process to occur. Both macrophages and T lymphocytes possess the capacity to regulate essential steps in the process of wound healing. The presence of macrophages is essential for the initiation and maintenance of wound fibroblast activity. T cells do not appear to be required for the initiation of the healing process, and healing can progress in the absence of T lymphocytes; however, the presence of an intact T cell immune system is essential for a normal outcome, indicating that the T cells probably provide a regulatory influence over macrophage-induced activities. Further research is still required into the interaction of these immune cells, their secretory products, and other wound elements before our understanding of the mechanism of wound healing is complete.

Introduction

The response of living tissue to injury forms the foundation of all surgical practice. Following tissue disruption, whether operative or traumatic, the major priorities of any organism are cessation of haemorrhage, prevention of infection and restoration of tissue integrity and function. In the specialization which has followed our anatomical and physiological evolution, we have lost the facility to regenerate organs and most tissues. Some lower forms of life, such as protozoa, can regenerate any part of their unicellular structures, while lower vertebrates retain the ability to regenerate an amputated limb or tail from totipotent cells; however, the mammalian system is only able to replace certain tissues, such as an epidermal defect, and must therefore repair damage to other structures by scar formation.


The process by which tissue repair takes place is termed wound healing and is comprised of a continuous sequence of inflammation and repair, in which epithelial, endothelial, inflammatory cells, platelets and fibroblasts briefly come together outside their normal domains, interact to restore a semblance of their usual discipline and having done so resume their normal function.

The process of wound repair differs little from one kind of tissue to another and is generally independent of the form of injury. Although the different elements of the wound healing process occur in a continuous, integrated manner, it is convenient to divide the overall process into three overlapping phases and several natural components for descriptive purposes (Fig. 1).

Inflammatory Phase (Day 0-5)

The healing response is initiated at the moment of injury. Surgical or traumatic wounds disrupt the tissue architecture and cause haemorrhage. Initially, blood fills the wound defect and exposure of this blood to collagen in the wound leads to platelet degranulation and activation of Hageman factor [1]. This in turn sets into motion a number of biological amplification systems including the complement kinin and clotting cascades and plasmin generation. These serve to amplify the original injury signal and lead not only to clot formation, which unites the wound edges, but also to the accumulation of a number of mitogens and chemoattractants at the site of wounding [2].

Production of both kinins and prostaglandins leads to vasodilatation and increased small vessel permeability in the region of the wound [3]. This results in oedema in the area of the injury and is responsible for the pain and swelling which occurs early after injury. Within 6 h, circulating immune cells start to appear in the wound. Polymorphonuclear leucocytes (PMN) are the first blood leucocytes to enter the wound site. They initially appear in the wound shortly after injury and subsequently their numbers increase steadily, peaking at 24-48 h [4]. Their main function appears to be phagocytosis of the bacteria which have been introduced into the wound during injury. The presence of PMN in the wound following injury does not appear to be essential in order for normal wound healing to take place [5, 6], with healing proceeding normally in their absence provided that bacterial contamination has not occurred. In the absence of infection, PMN have a relatively short life span in the wound and their numbers decrease rapidly after the third day [7].

The next cellular, immune element to enter the wound are macrophages. These cells are derived from circulating monocytes by a combination of migration and chemotaxis. They first appear within 48-96 h post-injury and reach a peak around the third day post-injury [4]. These macrophages have a much longer life span than the PMN and persist in the wound until healing is complete. Their appearance is followed somewhat later by T lymphocytes, which appear in significant numbers around the fifth day post-injury, with peak numbers occurring about the seventh day after injury. In contrast to PMN, the presence and activation of both macrophages and lymphocytes in the wound is critical to the progress of the normal healing process [8, 9].

Macrophages just like neutrophils phagocytose and digest pathological organisms and tissue debris. In addition, macrophages release a plethora of biologically active substances. Many of these substances facilitate the recruitment of additional inflammatory cells and aid the macrophage in tissue decontamination and debridement; in addition growth factors and other substances are also released which are necessary for the initiation and propagation of granulation tissue formation. These intercellular transmitters are known collectively as cytokines.

Proliferative Phase (Day 3-14)

In the absence of significant infection or contamination the inflammatory phase is short, and after the wound has been successfully cleared of devitalized and unwanted material it gives way to the proliferative phase of healing. The proliferative phase is characterized by the formation of granulation tissue in the wound. Granulation tissue consists of a combination of cellular elements, including fibroblasts and inflammatory cells, along with new capillaries embedded in a loose extra cellular matrix of collagen, fibronectin and hyaluronic acid. Fibroblasts first appear in significant numbers in the wound on the third day post-injury and achieve peak numbers around the seventh day [4]. This rapid expansion in the fibroblast population at the wound site occurs via a combination of proliferation and migration [10]. Fibroblasts are derived from local mesenchymal cells, particularly those associated with blood vessel adventitia [11], which are induced to proliferate and attracted into the wound by a combination of cytokines produced initially by platelets and subsequently by macrophages and lymphocytes (Table 1). Fibroblasts are the primary synthetic element in the repair process and are responsible for production of the majority of structural proteins used during tissue reconstruction. In particular, fibroblasts produce large quantities of collagen, a family of triple-chain glycoproteins, which form the main constituent of the extracellular wound matrix and which are ultimately responsible for imparting tensile strength to the scar. Collagen is first detected in the wound around the third day post-injury [12, 13], and thereafter the levels increase rapidly for approximately 3 weeks. It then continues to accumulate at a more gradual pace for up to 3 months post wounding [10]. The collagen is initially deposited in a seemingly haphazard fashion and these individual collagen fibrils are subsequently reorganized, by cross-linking, into regularly aligned bundles oriented along the lines of stress in the healing wound. Fibroblasts are also responsible for the production of other matrix constituents including fibronectin, hyaluronic acid and the glycosaminoglycans [14]. The process of fibroblast proliferation and synthetic activity is known as fibroplasia.

Revascularization of the wound proceeds in parallel with fibroplasia. Capillary buds sprout from blood vessels adjacent to the wound and extend into the wound space. On the second day post-injury, endothelial cells from the side of the venule closest to the wound begin to migrate in response to angiogenic stimuli. These capillary sprouts eventually branch at their tips and join to form capillary loops, through which blood begins to flow. New sprouts then extend from these loops to form a capillary plexus [15, 16]. The soluble factors responsible for angiogenesis remain incompletely defined. It appears that angiogenesis occurs by a combination of proliferation and migration. Putative mediators for endothelial cell growth and chemotaxis include cytokines produced by platelets, macrophages and lymphocytes in the wound [17, 18], low oxygen tension [19], lactic acid [20] and biogenic amines [21]. Of the potential cytokine mediators of neovascularization basic fibroblast growth factor (bFGF), acidic FGF (aFGF), transforming growth factors-a and b (TGF-a and -b) and epidermal growth factor (EGF) have all been shown to be potent stimuli for new vessel formation [22-24]. FGF, in particular, has been shown to be a potent inducer of in vivo neovascularization [25, 26].

While these events are proceeding deep in the wound, restoration of epithelial integrity is taking place at the wound surface. Re-epithelialization of the wound begins within a couple of hours of the injury. Epithelial cells, arising from either the wound margins or residual dermal epithelial appendages within the wound bed, begin to migrate under the scab and over the underlying viable connective tissue. The epidermis immediately adjacent to the wound edge begins thickening within 24 h after injury. Marginal basal cells at the edge of the wound loose their firm attachment to the underlying dermis, enlarge and begin to migrate across the surface of the provisional matrix filling the wound. Fixed basal cells in a zone near the cut edge undergo a series of rapid mitotic divisions, and these cells appear to migrate by moving over one another in a leapfrog fashion until the defect is covered [27, 28]. Once the defect is bridged, the migrating epithelial cells loose their flattened appearance, become more columnar in shape and increase in mitotic activity. Layering of the epithelium is re-established and the surface layer eventually keratinized [29]. Reepithelialization is complete in less than 48 h in the case of approximated incised wounds, but may take substantially longer in the case of larger wounds where there is a significant tissue defect. If only the epithelium is damaged, such as occurs in split thickness skin graft donor sites, then repair consists primarily of re-epithelization with minimal or absent fibroplasia and granulation tissue formation. The stimuli for re-epithelization remain incompletely determined, but it appears that the process is mediated by a combination of loss of contact inhibition, exposure of constituents of the extracellular matrix, particularly fibronectin [30], and by cytokines produced by immune mononuclear cells [31]. EGF, TGF-b, bFGF, platelet-derived growth factor (PDGF) and insulinlike growth factor-l (IGF-l) in particular, have been shown to promote epithelialization [32].

Maturation Phase (Day 7 to I Year)

Almost as soon as the extracellular matrix is laid down, its reorganization begins. Initially, the extracellular matrix is rich in fibronectin, which forms a provisional fibre network. This serves not only as a substratum for migration and ingrowth of cells, but also as a template for collagen deposition by fibroblasts [33]. There are also significant quantities of hyaluronic acid and large molecular weight proteoglycans present, which contribute to the gel-like consistency of the extracellular matrix and aid cellular infiltration. Collagen rapidly becomes the predominant constituent of the matrix. The initially randomly distributed collagen fibres become cross-linked and aggregated into fibrillar bundles, which gradually provide the healing tissue with increasing stiffness and tensile strength [34]. After a 5-day lag period, which corresponds to early granulation tissue formation and a matrix largely composed of fibronectin and hyaluronic acid, there is a rapid increase in wound breaking strength due to collagen fibrogenesis. The subsequent rate of gain in wound tensile strength is slow, with the wound having gained only 20% of its final strength after 3 weeks. The final strength of the wound remains less than that of uninjured skin, with the maximum breaking strength of the scar reaching only 70% of that of the intact skin [34].

This gradual gain in tensile strength is due not only to continuing collagen deposition, but also to collagen remodelling, with formation of larger collagen bundles [35] and alteration of intermolecular crosslinking [36]. Collagen remodelling during scar formation is dependent on both continued collagen synthesis and collagen catabolism. The degradation of wound collagen is controlled by a variety of collagenase enzymes, and the net increase in wound collagen is determined by the balance of these opposing mechanisms. The high rate of collagen synthesis within the wound returns to normal tissue levels by 6-12 months [37], while active remodelling of the scar continues for up to 1 year after injury and indeed appears to continue at a very slow rate for life.

As remodelling progresses, there is a gradual reduction in the cellularity and vascularity of the reparative tissue which results in the formation of a relatively avascular and acellular collagen scar. Grossly this can be observed as a reduction in erythema associated with the earlier scar and some reduction in the scar volume, resulting in a pale thin scar. This is normally a desirable feature of healing; however, in some cases shrinkage of the scar may give rise to an undesirable reduction in skin mobility resulting in contracture.

Wound contraction, i. e. inward movement of the wound edge, is a further important element in the healing process and should be distinguished from contracture. Sharply incised wounds without significant tissue loss, approximated early after injury, heal rapidly without the need for significant reduction in the wound volume. Such wounds are described as having healed by primary intention. Large wounds, however, particularly those associated with significant tissue loss, heal by secondary intention, with granulation tissue gradually filling the defect and epithelization proceeding slowly from the wound edges. Contraction of the wound edges can lead to a significant reduction in the quantity of granulation tissue required to fill the wound defect and a reduction in the area requiring reepithelization, with a consequent reduction in scar volume. Contraction is only undesirable where it leads to unacceptable tissue distortion and an unsatisfactory cosmetic result. Although contraction normally accounts for a larger part of overall wound closure in looseskinned animals, it still accounts for a significant proportion of the healing process in man, particularly in areas where the skin is not tightly bound down to underlying structures, such as on the back, neck and forearms. Initially following injury, where the wound edges are not approximated, there is a slight retraction of the wound edges due to the release of normal elastic tension in the skin, with a resultant increase in wound volume. The wound area starts to decrease rapidly from the third day onwards. While this is due in part to reepithelization, the main reason is an inward movement of the uninjured skin edges. Wound contraction usually begins around the fifth day postwounding and is complete by 12-15 days after wounding [38-40]. Fibroblasts within the wound appear to be responsible for providing the force for this contractile activity [41]. It was initially felt that specialized fibroblasts called myofibroblasts provided the motive force for wound contraction via a musclelike cell contraction [42-44]. More recent studies reveal that wound contraction occurs as a result of an interaction between fibroblast locomotion and collagen reorganization [41, 45]. The contraction is thought to be mediated via the attachment of collagen fibrils to cell surface receptors [46], with the resulting tractional forces generated by cell motility bringing the attached collagen fibrils closer together and eventually compacting them [47].

The regulation of wound contraction remains poorly defined. Information regarding the effects of specific cytokines on contraction is limited and often conflicting. TGF-b has been found to promote contraction even in the absence of serum [48, 49]; PDGF has also been found to either increase contraction [50] or have no effect [49], while both FGF and EGF have been found by different authors to either have no effect or cause a moderate enhancement of contraction [48-50].

Scar Formation

As mentioned previously, the process of wound healing is essentially similar in all tissues and is relatively independent of the mode of injury; however, slight variation in the relative contribution of the different elements to the overall result may occur. The final product of the healing process is a scar. This relatively avascular and acellular mass of collagen serves to restore tissue continuity, strength and function. Delays in the healing process cause the prolonged presence of wounds, while abnormalities of the healing process may lead to abnormal scar formation. Successful completion of wound healing may not always yield the desired clinical result, particularly where the final cosmetic appearance of the scar is of primary importance.

The Role of Macrophages

Macrophages appear to have a dual role at the wound site. Initially, they participate in the inflammatory and debridement process, superceding the PMN as the major wound phagocyte, and later, they play a regulatory role in the mediation of the fibroblastic phase of healing. It is this latter role which is crucial to the success of the wound healing process. In the classic studies of Leibovich and Ross, a combination of systemic hydrocortisone to induce systemic monocytopenia and local antimacrophage serum for local elimination of tissue macrophages resulted in a significant impairment of wound debridement and fibroplasia in guinea pigs [8]. Wound fibrin levels were elevated and clearance of fibrin, neutrophils erythrocytes and other miscellaneous debris from the wound was delayed in treated animals. In addition, there was both a delay in the appearance of fibroblasts in the wound and in the subsequent rate of expansion of the wound fibroblast population [51]. Recently, we have shown that in vivo macrophage depletion by parenteral administration of macrophagespecific monoclonal antibodies results in a significant reduction in both wound breaking strength and wound collagen deposition (A. Barbul, unpublished data).

Further evidence for macrophage involvement in the regulation of wound healing is provided by the findings that intradermal injection of allogeneic macrophages increases both collagen synthesis and wound breaking strength in 8day-old rat skin wounds [52], while injection of wound macrophages into rabbit corneas induces angiogenesis and scar formation [53, 54].

Activated macrophages are capable of influencing many aspects of wound healing, including the proliferative [51] and synthetic activities of fibroblasts [53] and induction of neovascularization [55, 56].

Macrophages mediate their effects on other wound elements via the release of monokines. These intracellular transmitters are capable of regulating fibroblast (Table 1) and endothelial function.

Of the cytokines produced by activated macrophages, TGF-b [57, 58], TNF-a and interleukin-1 (IL-1) [59] have been detected in significant quantities in the extracellular fluid at the site of healing. Topical application of TGF-b [60, 61] and PDGF [62] to healing wounds results in significantly enhanced wound breaking strength. Application of TGF-b at the time of wounding in rats with steroid-induced monocytopenia results in a wound breaking strength not significantly different from non-steroid-treated rats; this effect is not seen with topical PDGF application. Neither treatment resulted in an increase in the number of macrophages in the wound. TGF-b appeared to act directly on the fibroblasts, inducing type 1 procollagen gene expression, while the effects of PDGF appear to be mediated indirectly, possibly by attracting futher macrophages and inducing the release of other monokines. This hypothesis is supported by observations in models of impaired healing. Following total body irradiation, which induces monocytopenia with selective sparing of skin tissue, there is a significant decrease in wound breaking strength. Topical treatment with TGF-b [63], but not PDGF [64], results in significantly increased wound strength. By contrast, following megavolt electron beam surface irradiation to the skin surface, which impairs skin fibroblasts but spares the bone marrow, TGF-b [64] had no effect on subsequent healing, while PDGF [63] led to a 50 % increase in wound breaking strength, associated with an increase in the number of macrophages and fibroblasts in the healing wound.

IL-1, another macrophage product, has been shown to inhibit collagen synthesis in subcutaneously implanted sponges in rats [65]. While TNF-a has been found to have no effect on wound collagen deposition [66] when administered alone, it acted synergistically with PDGF and inhibited the effects of TGF-b on collagen deposition. However, administration of TGF-a into polyvinyl alcohol (PVA) sponges induces increased collagen deposition, an effect which is abrogated by simultaneous administration of indomethacin. This indicates a possible indirect inflammatory mode of action for TNF-a. On the other hand, TNF-a antibody administration, which blocks the effects of TNF-a, also induces an increase in collagen deposition [67] (Fig. 2). This data is consistent with a direct inhibitory effect of TNF-a on fibroblast synthetic activity; however, this effect is masked by its strong pro-inflammatory and macrophage activating effects.

In addition, topical application of bFGF results in an increase in wound breaking strength, collagen accumulation [68] and granulation tissue accumulation [69].

The Role of Lymphocytes

Evidence supporting a central role for T lymphocytes in the control of wound healing is provided by studies which examine the in vivo effects of alternate forms of T cell manipulation on various parameters of healing. Administration of agents known to enhance T lymphocyte function, such as growth hormone [70], vitamin A [71] or arginine [72], leads to increases in wound breaking strength and collagen deposition, while agents which suppress T lymphocyte function, such as steroids [73], retinoic acid, citral and cyclosporin A [74], markedly impair wound healing. Modification of T lymphocyte function by adult thymectomy, which prevents the induction of T suppressor cells, causes an increase in wound maturation. This effect could be reversed by intraperitoneal placement of autologous thymic grafts in millipore chambers in thymectomized rats [75]. Conversely, administration of purified thymic hormones , thymulin (FTS), thymopoietin and thymosin fraction V (TF5) results in impaired wound healing as assessed by wound breaking strength and wound collagen deposition [76]. These data suggest that the thymus exerts an inhibitory effect on normal wound healing, possibly by enhancing T suppressor cell activation following injury.

Direct evidence for lymphocyte involvement in the control of wound healing is provided by studies examining the effects of in vivo lymphocyte depletion of wound healing parameters. Global T cell depletion causes marked diminution in wound breaking strength and in the hydroxyproline content of subcutaneously implanted PVA sponges, used as an index of wound reparative collagen deposition [9, 77, 78]. Selective depletion of the T suppressor/cytotoxic lymphocyte subset causes marked enhancement of wound healing at 2 and 4 weeks post wounding. These findings suggests a possible role for the T suppressor/cytotoxic lymphocyte subset in the overall down-regulation of wound healing activity. It might be expected that the T helper/effector cells would promote such activity. Selective depletion of this subset, however, has no effect on either wound breaking strength or wound collagen deposition [9]. Simultaneous depletion of T helper/effector and T suppressor/cytotoxic T cells leads to significant increases in both wound breaking strength and collagen synthesis [77]. This suggests that an incompletely characterized T cell population bearing the T cell marker, but neither the T helper nor the T suppressor antigenic determinant, is responsible for the promotion of wound healing, since its deletion impairs wound healing.

Further support for these findings is provided by work in the congenitally athymic nude mouse [79]. These animals have a profoundly impaired T cell dependent immune system and display significantly enhanced wound breaking strength and collagen deposition in response to injury when compared to normal thymus-bearing animals. Administration of the anti-T cell monoclonal antibody to these athymic animals in order to deplete the small numbers of extra-thymically derived T cells present had no effect on either wound healing parameter but confirmed the previously observed significant decreases in wound breaking strength and hydroxyproline deposition when administered to normal thymus-bearing control mice. T cell reconstitution of nude mice, by injection of syngeneic T lymphocytes, resulted in significant decreases in wound breaking strength towards the levels observed in normal controls (Fig. 3).

Lymphocytes exert many of their effects via cytokines (Table 1). In vitro studies have shown that lymphokines are capable of modulating many fibroblast functions, including migration, replication and collagen synthesis. Some, such as TGF-b, lymphotoxin or g-interferon (IFN-g) are well characterized, while many other proteins which can modulate in vitro fibroblast activity have not been fully characterized. Both inhibitory and stimulatory lymphokines have been described for many fibroblast functions [80]; however, exactly how these various signals interact in vivo is presently unknown.

As mentioned previously, the presence of a number of potential regulators of healing has been confirmed in vivo at the wound site. The presence of biologically active TGF-b [57] and IL-6 has been shown in early wounds, although neither IL-2, IL-3 or IL-4 could be detected [59]. A similar model failed to detect either IFN-g or TGF-b at 10 days post-injury [81]. The effects of TGF-b administration on wound strength have already been mentioned [23]. Similar increases in both fresh and fixed wound breaking strength, with an associated rise in collagen deposition, were seen following administration of human recombinant IL-2 (60 000 U and 140 000 U/day) in rats [82] (Fig. 4). In contrast, the administration of IFN-g via subcutaneously implanted osmotic pumps in mice resulted in a decrease in the thickness and collagen content of the capsule which formed around the device [83].

References

1. Kaplan AP (1983) Hageman factor dependent pathways: mechanism of initiation and bradykinin formation. Fred Proc 42: 3123-3127

2. Knighton DR, Ciresi K, Fiegel VD (1990) Stimulation of repair in chronic nonhealing, cutaneous ulcers using platelet derived wound healing formula. Surg Gynecol Obstet 170: 56-60

3. Gorman RR (1978) Prostaglandins, thromboxanes and prostacyclins. Int Rev Biochem 48: 81

4. Ross R, Benditt EP (1961) Wound healing and collagen formation. I. Sequential changes in components of guinea pig skin wounds observed in the electron microscope. J Biophysiol Biochem Cytol 11: 677-700

5. Simpson DM, Ross R (1971) Effects of heterologous antineutrophil serum in guinea pigs: haematological and ultrastructural observations. Am J Pathol 65: 49-102

6. Simpson DM, Ross R (1972) The neutrophilic leucocyte in wound repair. A study with anti-neutrophil serum. J Clin Invest 51: 2009-2023

7. Willoughby DA (1970) Some views on the pathogenesis of inflammation. In: Montagna W, Bently JP, Dobson R (eds) The dermis. Advances in the biology of skin. Appelton-Century-Crofts, Norwalk, pp 221-230

8. Leibovich SJ, Ross R (1975) The role of the macrophage in wound repair. A study with hydrocortisone and antimacrophage serum. Am J Pathol 78: 71-100

9. Barbul A, Breslin JR, Woodyard JP, Wasserkrug HL, Efron G (1989) The effect of in vivo T helper and T suppressor lymphocyte depletion on wound healing. Ann Surg 209: 479-483

10. Clarke RA (1988) Overview and general considerations of wound repair. In: Clarke RAF, Henson PM (eds) The molecular and cellular biology of wound repair. Plenum, New York, pp 3-23

11. Ross R, Everett NB, Tyler R (1970) Wound healing and collagen formation. VI. The origin of the wound fibroblast studied in parabiosis. J Cell Biol 44: 645

12. Diegelmann RF, Rothkopf LC, Cohen IK (1975) Measurement of collagen biosynthesis during wound healing. J Surg Res 19: 239

13. Madden JW, Peacock EE (1969) Studies on the biology of collagen during wound healing I. Rate of collagen synthesis and deposition in cutaneous wounds of the rat. Surgery 64: 288

14. Kurkinen M, Vaheri A, Roberts PJ, Stenmam S (1980) Sequential appearance of fibronectin and collagen in experimental granulation tissue. Lab Invest 43: 47-51

15. Ausprunk DH, Folkman J (1977) Migration and proliferation of endothelial cells in preformed and newly formed blood vessels during tumor angiogenesis. Microvasc Res 14: 53-65

16. Burger PC, Chandler DB, Klintworth OK (1983) Corneal neovascularization as studied by scanning electron microscopy of vascular casts. Lab Invest 48: 169-180

17. Knighton DR, Silver IA, Hunt TK (1981) Regulation of wound healing angiogenesis - effect of oxygen gradient and inspired oxygen concentration. Surgery 90: 262-270

18. Harlan JA (1987) Consequences of leucocyte vessel wall interactions in inflammatory and immune reactions. Semin Thromb Hemost 13: 434-444

19. Remensnyder JP, Majno G (1969) Oxygen gradients in healing wounds. Am J Pathol 52: 301-319

20. Imre G (1964) Role of lactic acid. Br J Ophtalmol 48: 75-82

21. Zauberman H, Michaelson IC, Bergmann F, Mauric DM (1969) Stimulation of neovascularization in the cornea by biogenic amines. Exp Eye Res 8: 77-83

22. Schreiber AB, Winkler ME, Derynck R (1986) Transforming growth factor alpha: a more potent angiogenic mediator than epidermal growth factor. Science 232: 12501253

23. Roberts AB, Sporn MB, Assoian RK (1986) Transforming growth factor type b: rapid induction of fibrosis and angiogenesis in vivo and stimulation of collagen formation in vivo. Proc Natl Acad Sci USA 83: 4167-4171

24. Lynch SE, Colvin RB, Antoniades HN (1989) Growth factors in wound healing. Single and synergistic effects on partial thickness porcine skin wounds. J Clin Invest 84: 640-646

25. Gospodarowicz D, Cheng J, Lui GM, Baird A, Esch F, Bohlen P (1985) Angiogenic factor is related to fibroblast growth factor. Endocrinology 117: 2383-2391

26. Gospodarowicz D, Ferrara N, Schweigerer L, Neufeld G (1987) Structural characterization and biological function of fibroblast growth factor. Endocrinol Rev 8: 95-114

27. Winter GD (1962) Formation of the scab and role of epithelialization in the skin of young domestic pigs. Nature 193: 293-294

28. Stenn KS, Depalma L (1988) Re-epithelialization. In: Clarke RAF, Hensen PM (eds) The molecular and cellular biology of wound repair. Plenum, New York, pp 321-335

29. Johnson FR, McMinn RMH (1991) The cytology of wound healing of the body surface in mammals. Biol Rev 35: 364

30. Woodley DT, Bachman PM, O' Keefe EJ (1991) The role of matrix components in human keratinocyte re-epithelialization. In: Barbul A, Caldwell MD, Eaglstein WH et al. (eds) Clinical and experimental approaches to dermal and epidermal repair. Normal and chronic wounds. Wiley-Liss, New York, pp 129-140

31. Lynch SE (1991) Interreaction of growth factors in tissue repair. In: Barbul A, Caldwell MD, Eaglstein WH et al. (eds) Clinical and experimental approaches to dermal and epidermal repair. Normal and chronic wounds. Wiley-Liss, New York, pp 341-357

32. Lynch JB (1989) Enhancement of wound healing by topical treatment with epidermal growth factor. N EngI J Med 321: 76

33. McDonald JA, Quade BJ, Broekelmann TJ et al. (1987) Fibronectin's cell-adhesive domain and an amino terminal matrix assembly domain participate in its assembly into fibroblast pericellular matrix. J Biol Chem 262: 2957-2967

34. Levenson SM, Geever EF, Crowley LV, Oates JF, Bernard CW, Rosen H (1965) The healing of rat skin wounds. Ann Surg 161: 293-308

35. Kirscher CW, Shetler MR (1974) Collagen and mucopolysaccarides in the hypertrophic scar. Connect Tissue Res 3: 205-213

36. Bailey AJ, Bazin S, Simms TJ, LeLeus M, Nicholetis C, Delaunay A (1975) Characterization of the collagen of hypertrophic and normal scars. Biochim Biophys Acta 405: 412-421

37. Barnes MI, Morton LF, Bennet RC, Bailey AJ (1975) Studies in collagen synthesis in the mature dermal scar in the guinea pig. Biochem Soc 3: 917-920

38. Grillo HC, Potsaid MS (1961) Studies in wound healing. IV. Retardation of contraction by local X-irradiation, and observations relating to the origin of fibroblasts in repair. Ann Surg 154: 741-750

39. Peacock EE (1984) Wound repair. In: Wound repair. Saunders, Philadelphia, pp 38 - 55

40. Van Winkle W (1967) Wound contraction. Surg. Gyn. Obstet. 125: 131

41. Ehrlich HP (1988) The role of connective tissue matrix in wound healing. In: Barbul A, Pines E, Caldwell M, Hunt TK (eds) Growth factors and other aspects of wound healing: biological and clinical implications. Liss, New York, pp 243-258

42. Gabbiani G, Majno G (1971) Presence of modified fibroblasts in granulation tissue and their possible role in wound contraction. Experimentia 27: 549-550

43. Gabbiani G, Hirschel BJ, Ryan GB, Statkov PR, Majno G (1972) Granulation tissue as a contractile organ. J Exp Med 135: 719-734

44. Rudolph R, Vande Berg J, Pierce GF (1991) Changing concepts in myofibroblast function and control. In: Janssen H, Rooman R, Robertson JIS (eds) Wound healing. Petersfield, Wrightson, pp 103-119

45. Bell E, Ivarsson B, Merrill C (1979) 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: 1274-1278

46. Grinnell F, Lamke CR (1984) Reorganization of hydrated collagen lattices by human skin fibroblasts. J Cell Sci 66: 31

47. Wang SY, Merrill C, Bell E (1988) Effects of aging and long-term subcultivation on collagen lattice contraction and intra-lattice proliferation in three rat cell types. Mech Ageing Dev 44: 127

48. Finesmith TH, Broadley KN, Davidson JM (1990) Fibroblasts from wounds of different stages of repair vary in their ability to contract a collagen gel in response to growth factors. J Cell Physiol 144: 99-107

49. Montesano R, Orci L (1988) Transforming growth factor b stimulates collagen matrix contraction by fibroblasts: implication for wound healing. Proc Natl Acad Sci USA 85: 4894-4897

50. Clarke RAF, Folkevard JM, Hart CE, Murray MJ, McPherson JM (1989) Platelet isoforms of platelet-derived growth factor stimulates fibroblasts to contract collagen matrices. J Clin Invest 84: 1036-1040

51. Leibovich SJ, Ross R (1976) A macrophage-dependent factor that stimulates the proliferation of fibroblasts in vitro. Am J Pathol 84: 501-513

52. Casey W, Peacock EJ, Chvapil M (1976) Induction of collagen synthesis in rats by transplantation of allogeneic macrophages. Surg Forum 27: 53

53. Hunt TK, Knighton DR, Thakral KK, Goodson WH, Andrews WS (1984) Studies on inflammation and wound healing: angiogenesis and collagen synthesis stimulated in vivo by resident and activated wound macrophages. Surgery 96: 48-54

54. Clarke RA, Stone RD, Leung DYK, Silver I, Hohn DC, Hunt TK (1976) Role of macrophages in wound healing. Surg Forum 27: 16-18

55. Greenberg GB, Hunt TK (1978) The proliferation response in vitro of vascular endothelial and smooth muscle cells exposed to wound fluid and macrophages. J Cell Physiol 97: 353-360

56. Polverini PJ, Cotran RS, Gimbrone MA, Unanue ER (1977) Activated macrophages induce vascular proliferation. Nature 269: 804-806

57. Cromack DT, Sporn MB, Roberts AB, Merino MJ, Dart LL, Norton JA (1987) Transforming growth factor b levels in rat wound chambers. J Surg Res 42: 622-628

58. Rotatori DS, Caffee HH, Copeland EM, Schultz SS (1990) Growth factors are present in human wound fluid. Surg Forum 41: 627-629

59. Ford HR, Hoffman R, Wing EJ, Magee DM, McIntyre L, Simmons RL (1989) Wound cytokines in the sponge matrix model. Arch Surg 124: 1422-1428

60. Mustoe TA, Pierce GF, Thomason A, Gramates P, Sporn MB, Deuel TF (1987) Accelerated healing of incisional wounds in rats induced by transforming growth factor. Science 237: 1333-1336

61. McGee, Broadley KN, Buckley A et al. (1989) Recombinant transforming growth factor-b accelerates incisional wound healing. Curr Surg 46: 103-110

62. Pierce GF, Mustoe TA, Senior RM et al. (1988) In vivo incisional wound healing augmented by platelet-derived growth factor and recombinant C-SIS gene homodimeric proteins. J Exp Med 167: 974-987

63. Mustoe TA, Purdy J, Gamates P, Deuel TF, Thomason A, Pierce GF (1989) Reversal of impaired wound healing in irradiated rats by platelet-derived growth factor-b. Am J Surg 1598: 345-350

64. Cromack D, Purdy J, Porras-Reyes B, Mustoe TA (1990) Acceleration of tissue repair by transforming growth factor (TGF-b): elucidation of in vivo mechanism of action by selective radiotherapy impaired healing. Surg Forum 41: 630-631

65. Laato M, Heino J (1988) Interleukin 1 modulates collagen synthesis by rat granulation tissue cells both in vivo and in vitro. Experientia 44: 32-29

66. Steenfos HH, Hunt TK, Scheuenstuhl H, Goodson WH (1989) Selective effects of tumor necrosis factor-alpha on wound healing in rats. Surgery 106: 171-176

67. Regan MC, Kirk SJ, Hurson M, Sodeyama M, Wasserkrug HL, Barbul A (1991) Tumor necrosis factor-a inhibits in vivo collagen synthesis at the wound site. Surgery 113: 173-177

68. McGee GS, Davidson JM, Buckley A et al. (1988) Recombinant basic fibroblast growth factor accelerates wound healing. J Surg Res 45: 145-153

69. Sprugel KH, McPherson JM, Clowes AW, Ross R (1988) The effects of different growth factors in subcutaneous wound chambers. In: Barbul A, Pines E, Caldwell M, Hunt TK (eds) Growth factors and other aspects of wound healing: biological and clinical implications. Liss, New York, pp 77-91

70. Prudden JF, Nishihara G, Ocamp L (1958) Studies on growth hormone. III. The effect on wound tensile strength and marked postoperative anabolism induced with growth hormone. Surg Gynecol Obstet 107: 481-482

71. Ehrlich HP, Hunt TK (1968) Effect of cortisone and vitamin A on wound healing. Ann Surg 167: 324-326

72. Barbul A, Rettura G, Levenson SM (1977) Arginine: a thymotrophic and wound promoting agent. Surg Forum 28: 101-103

73. Sandberg N (1964) Time relationship between administration of cortisone and wound healing in rats. Acta Chir Scand 127: 446-455

74. Fishel RS, Barbul A, Wasserkrug HL (1983) Cyclosporin A impairs wound healing in rats. J Surg Res 34: 573-575

75. Barbul A, Sito D, Rettura G, Levinson SM, Seifter E, Efron G (1982) Thymic inhibition of wound healing: abrogation by adult thymectomy. J Surg Res 32: 338-342

76. Barbul A, Shawe T, Frankel H, Efron JE, Wasserkrug HL (1989) Inhibition of wound repair by thymic hormones. Surgery 106: 373-377

77. Efron JE, Frankel HL, Lazarou SA, Wasserkrug HL, Barbul A (1990) Wound healing and T lymphocytes. J Surg Res 48: 460-463

78. Peterson JM, Barbul A, Breslin RJ, Wasserkrug HL, Efron G (1987) Significance of T lymphocytes in wound healing. Surgery 102: 300-305

79. Barbul A, Shaw T, Rotter SM, Efron JE, Wasserkrug HL, Badaway SB (1989) Wound healing in nude mice: a study on the regulatory role of lymphocytes in fibroplasia. Surgery 105: 764-769

80. Freundlich B, Bamalaski JS, Neilson E, Jimenes SA (1986) Regulation of fibroblast proliferation and collagen synthesis by cytokines. Immunol Today 7: 303-307

81. Lazarou SA, Barbul A, Wasserkrug HL, Efron G (1990) The wound is a possible source of post-traumatic immunosuppression. Arch Surg 124: 1429-1431

82. Barbul A, Knud-Hansen J, Wasserkrug HL, Efron G (1986) Interleukin 2 enhances wound healing in rats. J Surg Res 40: 315-319

83. Granstein RD, Murphy GF, Margolis RJ, Byrne MH, Amento EP (1987) Gammainterferon inhibits collagen synthesis in vivo in the mouse. J Clin Invest 79: 12541258