Modified cotton gauze dressings that selectively absorb neutrophil elastase activity in solution.

Dressings for chronic human wounds have been aimed at protection, removal of exudate, and improved appearance. However since the time of ancient Greece wound care and dressing strategies have primarily relied on empiricism. Recent studies have shown that chronic wounds contain high levels of tissue and cytokine destroying proteases including collagenase and neutrophil elastase. Therefore we sought to develop an effective wound dressing that could absorb elastase through affinity sequestration. Cotton gauze was modified by oxidation, phosphorylation, and sulfonation to enhance elastase affinity by ionic or active site uptake. Type VII absorbent cotton gauze was oxidized to dialdehyde cotton which was subsequently converted in part to the bisulfite addition product. Gauze preparations were also phosphorylated and carboxymethylated. Modified cotton gauzes were compared with untreated gauze for reduction of elastase activity in buffered saline. Solutions of elastase that were soaked in oxidized, sulfonated, and phosphorylated cotton gauze showed reduced elastase activity. The initial velocities (v(o)) and turnover rates of elastase showed significant decreases compared with solutions taken from untreated gauze. The reduction in enzyme activity with dialdehyde cotton gauze was confirmed in solution by determining elastase inhibition with dialdehyde starch. The dialdehyde cotton gauze also decreased elastase activity in human wound fluid in a dose response relation based on weight of gauze per volume of wound fluid. Absorbency, pH, air permeability and strength properties of the modified gauze were also compared with untreated cotton gauze. This report shows the effect of reducing elastase activity in solution with cotton containing aldehydic or negatively charged cellulose fibers that may be applicable to treatment modalities in chronic wounds.

Proinflammatory cytokines differentially regulate hyaluronan synthase isoforms in fetal and adult fibroblasts.

BACKGROUND/PURPOSE: Fetal wound healing is a relatively scarless process that occurs in an hyaluronan-rich environment. Understanding the regulation of hyaluronan expression may provide insight into the process of fetal repair. Therefore, the purpose of this study was to compare the regulation of hyaluronan and hyaluronan synthase transcripts by the proinflammatory cytokines interleukin-1beta (IL-1beta) and tumor necrosis factor-alpha (TNF-alpha) in human adult and fetal fibroblasts. METHODS: Hyaluronan deposited in the medium of untreated fibroblasts or fibroblasts treated with either IL-1beta or TNF-alpha was determined by an assay utilizing iodine I 125-hyaluronan binding protein. HAS transcript levels were compared in using a ribonuclease protection assay. RESULTS: IL-1beta induced an increase in hyaluronan accumulation by both fetal and adult fibroblasts. In contrast, TNF-alpha induced higher levels of hyaluronan only in fetal fibroblasts. HAS-2 and HAS-3 transcript levels were constitutively expressed by both fetal and adult fibroblasts. Proinflammatory cytokines induced a differential increase in HAS-1 and HAS-3 transcript levels. CONCLUSIONS: Differential regulation was observed in hyaluronan accumulation and for HAS transcript levels in fetal and adult dermal fibroblasts. The muted response of fetal fibroblasts to cytokines may be relevant to the minimal inflammation associated with fetal repair.

Myofibroblast induction with transforming growth factor-beta1 and -beta3 in cutaneous fetal excisional wounds.

BACKGROUND/PURPOSE: In a noncontractile fetal rabbit model, the authors recently have shown the induction of excisional wound contraction with sustained-release cellulose implants formulated with transforming growth factor (TGF)-beta. The purpose of this study was to test the hypothesis that the excisional wound contraction in this model is associated with the induction of myofibroblasts in the surrounding dermis, demonstrated by the presence of alpha-smooth muscle actin. METHODS: Cellulose discs were formulated with either 1.0 microg of TGF-beta1 (n = 6); 1.0 microg of TGF-beta3 (n = 9); 10 microg of TGF-beta3 (n = 6); or their carrier protein, bovine serum albumin (BSA; n = 9), for sustained-release over 5 days. Each disc was implanted into a subcutaneous pocket on the back of a fetal New Zealand White rabbit in utero on day 24 of gestation (term, 31 days). A full-thickness, 3-mm excisional wound (7.4 mm2) was then made next to the implanted cellulose disc. All fetuses were harvested at 3 days. The amount of alpha-smooth muscle (SM) actin in the dermis around the implants and wounds was determined using immunohistochemical techniques. RESULTS: Excisional wounds exposed to 1.0 microg of TGF-beta1 (5.6+/-2.0 mm2), 1.0 microg of TGF-beta3 (6.9+/-1.0 mm2), and 10 microg of TGF-beta3 (2.7+/-1.0 mm2) were significantly smaller when compared with the BSA control group (12.8+/-1.1 mm2; P<.05). Furthermore, there was a significant increase in staining for alpha-SM actin in the TGF-beta1 (1.8+/-0.5) and 10 microg TGF-beta3 (2.8+/-0.2) groups in comparison with the scant staining in the BSA control group (0.5+/-0.2; P<.05). CONCLUSIONS: TGF-beta1 and -beta3 induce alpha-SM actin and contraction of cutaneous excisional wounds in a fetal noncontractile model. This model of inducible cutaneous excisional wound contraction may be useful in further determining the role of the myofibroblast in wound contraction and the physiology underlying this poorly understood aspect of wound healing.

Method to analyze collagenase and gelatinase activity by fibroblasts in culture.

The net amount of collagen produced and deposited by fibroblasts in cell culture is determined by the rate of collagen synthesis as well as the rate of collagen degradation. Although collagen synthesis can be analyzed by several techniques, it is more difficult to measure collagen degradation. Breakdown of collagen depends upon the activity of a family of structurally and catalytically related mammalian enzymes termed matrix metalloproteinases (MMPs). Interstitial collagenase (MMP1) initiates the cleavage of fibrillar collagen, whereas gelatinases (MMP2 and MMP9) digest the denatured collagen fragments. A method has been developed to quantitate the activity of collagenase (MMP1) and gelatinase (MMP9) in conditioned medium from fibroblast cell cultures. The assay, which uses the fluorogenic substrate Dnp-Pro-Cha-Gly-Cys(Me)-His-AlaLys(Nma)NH2, is technically simple and amenable to high throughput analysis. Addition of specific inhibitors of the metalloproteinases allows for simultaneous measurement of both collagenase and gelatinase activity.

TGF-beta1 alters the healing of cutaneous fetal excisional wounds.

BACKGROUND/PURPOSE: In a number of species, fetal wound healing differs from the adult in the absence of inflammation, fibrosis, scar formation, and excisional wound contraction. The lack of inflammation also may explain the relative absence of any cytokine levels at the wound site, such as transforming growth factor (TGF)-beta, and therefore the unique characteristics of fetal wound healing. The authors hypothesized that exogenous TGF-beta1 would induce contraction, inflammation, fibrosis, and scar formation in cutaneous excisional wounds in the fetal rabbit. METHODS: Cellulose discs (3 mm in diameter) were formulated with either 1.0 microg TGF-beta1 (n = 6) or bovine serum albumin (BSA; n = 7), as a control, for sustained-release over 3 days. Each disc was implanted into the subcutaneous tissue on the backs of fetal New Zealand White Rabbits in utero on day 24 of gestation (term, 31 days). A full-thickness, 3-mm excisional wound (7.4 mm2) was then made next to the implanted cellulose disc. All wounds were harvested 3 days later. RESULTS: At harvest, the excisional wounds in the TGF-beta1 group had contracted (5.6 +/- 2.0 mm2), whereas those in the control group had expanded (13.5 +/- 1.2 mm2, P< .01). The surrounding dermis in the TGF-beta1 group had 16.3 inflammatory cells per grid block compared with 12.4 cells in the control group (not significant). In addition, a greater amount of fibrosis was induced by the TGF-beta1 implant (1.7 +/- 0.3) than the control implant (0.4 +/- 0.2) on a scale of 0 to 3, P < .01. In situ hybridization analysis showed an increase in procollagen type 1alpha1 gene expression in the surrounding dermis of the TGF-beta1 group (36.7 +/- 3.6 grains per grid block) compared with the control group (7.1 +/- 0.9 grains per grid block, P < .001). CONCLUSIONS: These results demonstrate that the cytokine TGF-beta1 can induce fetal excisional wounds to contract, stimulate fibrosis, and increase procollagen type 1alpha1 gene expression. These findings further suggest that the absence of TGF-beta1 atthe wound site may be responsible in part for the lack of a postnatal healing response.

MMP-8 is the predominant collagenase in healing wounds and nonhealing ulcers.

BACKGROUND: The initial cleavage of collagen by collagenase represents the rate-limiting step in the degradation of this central extracellular matrix protein. Chronic nonhealing ulcers, especially pressure ulcers, typically contain elevated levels of collagenolytic activity. However, there have been no detailed attempts to identify the source of these collagenases and their activity either in normal healing wounds or in chronic nonhealing ulcers. MATERIALS AND METHODS: Levels of the matrix metalloproteinases, MMP-1 and MMP-8, and the tissue inhibitor of matrix metalloproteinases, TIMP-1, were measured in fluids and tissues of healing human wounds and nonhealing ulcers by ELISA. Relative MMP-1 and MMP-8 levels were also analyzed by substrate preference in a functional assay. RESULTS: The patterns of the collagenases MMP-1 and MMP-8 in healing wounds were distinct, with MMP-8 appearing in significantly greater amounts than MMP-1. Chronic nonhealing ulcers were characterized by significantly higher levels of MMP-1 and MMP-8, and lower levels of TIMP-1, than in healing wounds. Levels of both MMP-1 and MMP-8 varied greatly in chronic ulcers, although MMP-8 was always the predominant collagenase present in these wounds. Interestingly, these collagenases were present almost exclusively in their inactive forms in healing wounds, whereas nonhealing ulcers possessed significant levels of the active forms of these enzymes. CONCLUSIONS: These results clearly demonstrate that the neutrophil-derived MMP-8 is the predominant collagenase present in normal healing wounds and suggest that overexpression and activation of this collagenase may be involved in the pathogenesis of nonhealing chronic ulcers. In addition, excessive collagenolytic activity in chronic ulcers is made possible, partly because of the reduced levels of the inhibitor, TIMP-1.

The proteolytic environment of chronic wounds.

A consistent feature of chronic leg and pressure ulcers is chronic inflammation associated with an elevated infiltration of neutrophils. Neutrophils and their proteases have been implicated in mediating the tissue damage associated with a variety of chronic inflammatory diseases. This review discusses our current understanding of the proteolytic enzymes found in chronic wounds and attempts to relate this information to the abundant presence of neutrophils. In addition, the implications that the proteolytic environment may have for current and future treatment strategies of chronic nonhealing wounds are discussed.

Dynamics of the matrix metalloproteinases MMP-1 and MMP-8 in acute open human dermal wounds.

Extracellular matrix degradation during dermal wound healing involves multiple levels of regulation by several enzymes of the matrix metalloproteinase family, their activators, and their inhibitors. This study tested the hypothesis that a temporal pattern of interstitial collagenase appearance occurs during normal dermal wound healing, with matrix metalloproteinase-8 originating from neutrophils appearing earlier than the fibroblast-derived matrix metalloproteinase-1. Open (6 mm) full-thickness dermal wounds, which were covered by transparent occlusive dressings, were made in healthy human volunteers (n = 20). Wound fluids from under the dressings were collected daily through day 8, and wound tissue biopsies were obtained on days 0, 2, 4, 14, and 28. Collagenases were extracted from homogenized tissue biopsies for analysis. Samples were analyzed for the presence of matrix metalloproteinase-1 and matrix metalloproteinase-8 by enzyme-linked immunosorbent assays and by collagenase activity assays using purified types I and III collagen as substrates. In addition, tissue inhibitor of metalloproteinases-1 and matrix metalloproteinase-1/tissue inhibitor of metalloproteinases-1 complexes in wound fluids were measured. Results showed a differential temporal pattern of matrix metalloproteinase-1 and matrix metalloproteinase-8 in wound exudates with peak levels of matrix metalloproteinase-8 occurring on day 4 and matrix metalloproteinase-1 peak levels on day 7. Maximal levels in tissue for both enzymes occurred on day 2. At all time points examined, levels of matrix metalloproteinase-8 were statistically higher than matrix metalloproteinase-1 (100-fold to 200-fold). Tissue inhibitor of metalloproteinases-1 levels declined over time, whereas levels of matrix metalloproteinase-1/tissue inhibitor of metalloproteinase-1 complexes increased to a plateau on day 7. This study provides new evidence implicating matrix metalloproteinase-8 as a major collagenase in healing human dermal wounds. It also shows a temporal pattern in the appearance of the matrix metalloproteinases, tissue inhibitor of metalloproteinase-1, and matrix metalloproteinase-1/tissue inhibitor of metalloproteinases-1 complexes, suggesting that a tightly regulated pattern of expression of matrix metalloproteinases and their inhibitors is essential for normal wound healing in humans.

Interleukin-1alpha and collagenase activity are elevated in chronic wounds.

Interleukin-1-alpha (IL-1alpha) is a member of a family of proinflammatory polypeptide mediators that has been shown in vitro to stimulate collagenase production. Collagenase is a proteolytic enzyme classified as one of the matrix metalloproteinases (MMP-1) that specifically recognizes and cleaves collagen. Therefore, the objective of this study was to compare the levels of these two proteins in chronic wounds as possible factors in the pathogenesis of chronic wounds. Fluids from 10 chronic wounds were collected before and after a 1-week treatment with a hydroactive dressing (Cutinova cavity). In addition, fluids were collected from 20 acute wounds for comparison. IL-1alpha and MMP-1 levels were quantified using sandwich ELISA. Collagenase activity was measured using a radiolabeled collagen as substrate. Clinically, the chronic wounds showed decreased area (-21.0 cm2) and reduced volume (-134.5 cm3) by 4 weeks after treatment with the hydroactive dressing. There were no significant differences in the protein concentrations between acute wound fluids (21.0 +/- 3.0 mg/ml) and chronic wound fluids before and after treatment with the hydroactive dressing (18.3 +/- 5.5 and 25.2 +/- 7.6 mg/ml, respectively). Levels of IL-1alpha in the acute wound fluids were low (0.019 pg/mg), whereas in the chronic wound fluid before treatment they had been significantly elevated (44.9 + 21.8 pg/mg). Following treatment with the hydroactive dressing, the IL-1alpha levels dropped to 10.3 + 3.3 pg/mg (p < 0.05). Collagenase activity was not detectable in acute wound fluid, elevated in pretreatment chronic wounds (12.9 + 3.4 units), and decreased in chronic wounds after treatment (11.4 + 3.3 units). This study correlated clinical healing of chronic wounds with biochemical changes in the ulcer microenvironment. As the chronic wounds began to heal, there was a significant decrease in the IL-1alpha levels and collagenase activity, thus suggesting that these two proteins may contribute to the lack of healing characteristic of chronic wounds.

Physiology of the chronic wound.

There is as yet no unified mechanism that explains the pathophysiology of every nonhealing wound. This is largely because many of the basic biological investigations in this area have only been undertaken seriously in the past few years and many phenomena remain unexplained. As new data become available, these theoretical models undoubtedly will continue to evolve. However, a clear understanding of the facts presently available should provide the clinician with a good scientific basis for rational clinical interventions. There is an urgent need to bridge the gap that often exists between laboratory research and clinical practice. A number of wound care practices, which have been shown to be ineffective or harmful, are still widely used. This includes the application of toxic wound cleansing agents, inappropriate use of topical antibiotics, and the practice of wet-to-dry dressings. The past 2 decades have witnessed an unprecedented proliferation of wound care products, few of which have be proven to be consistently superior to simpler and more cost-effective measures. For the foreseeable future, the search for the magic wound 'portions and lotions' probably will continue to revolve around topical growth factor and antiprotease therapy. However, such efforts will only come to fruition when we truly understand the pathophysiologic basis for abnormal wound healing.

Decreased expression of tissue inhibitor of metalloproteinase 1 in stunned myocardium.

Ultrastructural studies of stunned myocardium have shown disorganization and loss of extracellular collagen and increased collagenase activity early after ischemia and reperfusion. The interplay between matrix metalloproteinase 1 (MMP-1) and tissue inhibitor of metalloproteinase 1 (TIMP-1) regulates the turnover of cardiac extracellular matrix fibrillar collagens. However, the gene expression of MMP-1 and TIMP-1 in stunned myocardium is not known. Here, we determined whether altered expression of MMP-1 and TIMP-1 occurs in globally stunned hearts. An isolated nonworking rabbit heart preparation, perfused with a bovine erythrocyte suspension in modified Krebs solution, was used. Two groups were studied: the stunned group was subjected to 20 min of normothermic global ischemia followed by 120 min of normal reperfusion (n = 8), and the control group underwent 140 min of uninterrupted perfusion (n = 7). The developed pressures at the end of reperfusion for ischemic and control hearts were 67.0 +/- 2.73 and 83.1 +/- 1.52 mm Hg (P < 0. 006) respectively. Ribonuclease protection assays of total left ventricular RNA using riboprobes for MMP-1, TIMP-1, and 18S rRNA were performed. A significant decrease (twofold, P < 0.03) in TIMP-1 gene expression was found in the stunned hearts, while MMP-1 mRNA expression was unchanged. Thus, in early stunning, the decrease in TIMP-1 expression could tip the balance favoring enhanced metalloproteinase activity, promoting collagen turnover, and initiating extracellular matrix remodeling. This may contribute to delayed recovery from myocardial stunning.

The nitric oxide precursor L-arginine reduces expression of hyaluronan synthase in experimental vein bypass grafts.

BACKGROUND: The success of vascular bypass procedures is limited by the development of intimal hyperplasia (IH). The nitric oxide (NO) precursor, L-arginine (L-ARG) significantly reduces IH in both arteries and experimental vein grafts; however, the precise mechanism has yet to be elucidated. Hyaluronan synthase-1 (HAS-1) is one of the two enzymes believed to be responsible for making hyaluronan, a key component extracellular matrix composition. PURPOSE: To determine how L-ARG supplementation affects the gene expression of HAS-1 in experimental vein grafts. METHODS: Thirty-four male New Zealand white rabbits were divided into three groups: control (no operation, regular chow and water, n = 4); L-ARG supplemented (n = 15); and no L-ARG (n = 15). The latter two groups underwent a right interposition carotid bypass using jugular vein. Vein grafts were harvested at 7, 14, and 21 days after surgery. Ribonuclease protection assays were performed using 32P-labeled riboprobes for HAS-1 and 18S rRNA as an internal control and expressed as a ratio (HAS-1/rRNA). RESULTS: There was a significant rise in HAS-1 expression in the vein grafts 7 (1.57 +/- 0.5), 14 (0.7 +/- 0.2), and 21 days (2.82 +/- 0.7) after grafting compared to control (0.14 +/- 0.08) (P < 0.05). L-ARG-supplemented animals had a significant decrease in HAS-1 expression at 21 days (0.65 +/- 0.1) compared to nonsupplemented vein grafts (2.82 +/- 0.7) (P < 0.02). CONCLUSIONS: These results demonstrate for the first time a significant rise in HAS expression in the early experimental vein grafts. Furthermore, L-ARG supplementation significantly diminishes the expression of HAS at 21 days. These results may represent a potential mechanism by which augmentation of the L-ARG/NO pathway inhibits IH in experimental vein grafts and may ultimately provide for improved therapeutic interventions in alleviating IH.

Hypercholesterolemia alters the gene expression of novel components of the extracellular matrix in experimental vein grafts.

The success rate of vascular bypass procedures is limited by the development of intimal hyperplasia (IH). Hypercholesterolemia has been shown to accelerate IH in both arteries and experimental vein grafts; however the mechanism remains uncertain. Hyaluronic acid synthase (HAS-1) is a transmembrane enzyme responsible for the formation of hyaluronan; an important constituent of extracellular matrix (ECM). The integrin receptor for hyaluronan is CD-44. Both CD-44 and HAS-1 have been studied in the development of ECM of wounds but have yet to be examined in the ECM of IH within vein grafts. The purpose of this study was to determine if the expression of CD-44 and HAS-1 is increased during the early stages of IH and how cholesterol supplementation affects these genes. Forty white male New Zealand rabbits were divided into two groups: cholesterol supplemented (1% cholesterol chow) and noncholesterol supplemented. Each set of 20 rabbits was then divided into four additional groups (n = 5); a nonoperative group (control) and three operative groups that underwent a right interposition carotid bypass using jugular vein. Grafts were harvested at 3, 7, and 21 days after surgery for molecular studies and histology. Ribonuclease protection assays were performed using 32P-labeled riboprobes for HAS-1, CD-44, and 18s rRNA. Densitometric analysis is expressed as a ratio (riboprobe/rRNA). Cholesterol levels differed significantly between cholesterol supplemented and nonsupplemented groups (1419 +/- 130 mg/dl and 48 +/- 12 mg/dl) (p < 0.01). There was a significant increase in the expression of HAS-1 and CD-44 in the vein grafts compared to normal jugular vein. Cholesterol supplementation caused a further increase in CD-44 gene expression versus nonsupplemented vein grafts. These data demonstrate a role for CD-44 and HAS-1 transcription in vein graft intimal hyperplasia, which is further altered by cholesterol supplementation. Lastly, these results could explain differences seen in the development of IH with hypercholesterolemia and ultimately provide for improved therapies in alleviating this process.

Corticosteroids repress the interleukin 1 beta-induced secretion of collagenase in human intestinal smooth muscle cells.

BACKGROUND & AIMS: The cytokine interleukin (IL)-1 beta induces collagenase expression and inhibits collagen expression in human intestinal smooth muscle (HISM) cells. Corticosteroids cause transrepression of certain genes, including the collagenase gene. The aim of this study was to determine if corticosteroids repress the induction of collagenase expression and the inhibition of collagen secretion by IL-1 beta in HISM cells. METHODS: HISM cells were exposed to IL-1 beta (1-100 pmol/L) for 24 hours in the presence or absence of dexamethasone (10(-6) mol/L). Collagenase messenger RNA (mRNA) levels were determined by ribonuclease protection assay. Collagenase and tissue inhibitor of metalloproteinase protein secretion were determined by enzyme-linked immunosorbent assay of culture medium. Procollagen and collagen secretion were determined by polyacrylamide slab gel analysis of radiolabeled proteins in culture medium. RESULTS: A 30-fold induction of collagenase mRNA and collagenase protein secretion by IL-1 beta was completely abrogated by dexamethasone. Dexamethasone at 10(-6) mol/L also reduced basal levels of collagenase mRNA by 50% and blocked the IL-1 beta-induced inhibition of collagen secretion. CONCLUSIONS: Corticosteroids repress the collagenolytic action of IL-1 beta on HISM cells in vitro and therefore should promote healing in the inflamed intestine.

Inducible nitric oxide synthase expression in human vein grafts.

BACKGROUND: The patency of vascular reconstructive procedures is limited by the development of intimal hyperplasia (IH). Nitric oxide (NO) seems to be beneficial in abrogating this process. Currently, there is little information concerning inducible nitric oxide synthase (iNOS), the enzyme responsible for NO synthesis, and human vein grafts. The purpose of this study was to examine iNOS gene expression in human aortocoronary vein grafts (ACVG) and infrainguinal vein bypass grafts (IVG). METHODS: Nonthrombosed sections from ACVG (n = 5), IVG (n = 5), and control saphenous vein (SV; n = 4) were harvested and processed for RNA isolation. Quantitative reverse transcription-polymerase chain reaction (RT-PCR) was performed on samples using 32P radioactively end labeled primers. Glyceraldehyde-3-phosphate-dehydrogenase (GAPDH) was the internal control, and results were expressed as iNOS pmol/GAPDH pmol. RESULTS: There was a significant increase in the iNOS gene expression in the ACVG (0.049 +/- 0.01) when compared with IVG (0.019 +/- 0.001) or normal SV (0.011 +/- 0.002; P < or = 0.05). There was no significant difference between normal vein and the infrainguinal grafts. Sequencing of a fragment of the amplified 428 bp gene product confirmed 84% homology with the available gene bank human sequence. CONCLUSIONS: This study proves that iNOS is expressed in human vein bypass grafts. Additionally, there is a significant elevation of iNOS message in human ACVGs compared with IVG or normal SV. This difference may be the result of the unique vascular beds supplied by these grafts. Ultimately, manipulation of iNOS expression may lead to therapies to alleviate IH in these grafts.

Hyaluronic acid inhibits fetal platelet function: implications in scarless healing.

Platelets are important for the initiation of inflammation in adults, but the role of fetal platelets in fetal wound healing is unclear because fetal dermal wounds heal with a minimal inflammatory response and lack of excessive scarring. Because fetal tissue is abundant in glycosaminoglycans (GAGs), predominantly hyaluronic acid (HA), this study was designed to test the hypothesis that HA inhibits the reactivity of platelets and thus contributes to the minimal scarring characteristic of fetal tissue repair. Platelets were isolated from 10 fetal pigs at day 80 of gestation (term, 115 days) and exposed to 0.5 mg/mL of arachidonic acid, an agent shown in prior studies to evoke maximal aggregation and degranulation of fetal platelets. The ability of HA at 0.1 and 0.5 mg/mL to inhibit this response was determined. The presence of HA resulted in a dose-dependent reduction in platelet aggregation at 180 seconds (control, 99.7 +/- 0.3%; HA [0.1 mg/mL] 91.7 +/- 3.8%; and HA [0.5 mg/mL] 48.5 +/- 9.0%; P < .005 v control). The onset of aggregation was also significantly delayed by 0.5 mg/mL of HA (13.5 +/- 2.5 seconds) compared to control (2.9 +/- 0.7 seconds), P < .05. No significant diminution of platelet aggregation could be achieved by the addition of other GAGs at similar concentrations. HA also significantly impaired the release of platelet-derived growth factor (PDGF)-AB from fetal platelets. The authors conclude that HA, the predominant GAG in fetal dermal matrix, inhibits platelet aggregation and cytokine release. This inhibition of platelet aggregation and resultant inflammatory response may explain, in part, the minimal inflammation and scarless healing characteristic of fetal dermal repair.

Collagen induces cytokine release by fetal platelets: implications in scarless healing.

In previous studies the authors demonstrated that unlike adult platelets, fetal platelets respond poorly to collagen. When platelets make contact with the exposed collagen at the site of injury, the result is activation, aggregation, and degranulation with the release of cytokines and growth factors. This sequence of events is well characterized in adult wounds, which heal with an acute inflammatory response and dense scar formation. In sharp contrast, fetal dermal wounds heal without an acute inflammatory response and minimal scar formation. Therefore, the aim of this study was to test the hypothesis that collagen, abundant at the site of dermal injury, is a poor inducer of cytokine release by fetal platelets. This could explain, in part, the minimal inflammation accompanying fetal dermal wound healing. Platelet suspensions from six fetal Yorkshire swine at day 80 of gestation (term, 114 days) were exposed to either arachidonic acid, 0.5 mg/mL, collagen, 0.19 mg/mL, or saline. The release into plasma of transforming growth factor-beta (TGF-beta 1), and platelet-derived growth factor (PDGF)-AB effected by these agents was determined by enzyme-linked immunosorbent assays. Transmission electron microscopy (TEM) was used to correlate the biochemical findings with ultrastructural changes and showed that arachidonate-treated platelets were aggregated and devoid of granules. In contrast, collagen-treated platelets had undergone conformational changes but showed only a moderate change in the quantity and homogeneity of their secretory granules compared with saline-treated controls. There was a significant increase in TGF-beta 1 release into plasma after treatment with collagen (6.64 +/- 0.36 ng/mL) and arachidonate (7.64 +/- 0.77 ng/mL) compared with saline (4.74 +/- 0.36 ng/mL), P < .05. Likewise, PDGF-AB release was significantly higher after collagen (0.22 +/- 0.02 ng/mL) and arachidonate treatment (0.44 +/- 0.04 ng/mL) compared with saline (0.09 +/- 0.02 ng/ mL), P < .05. The authors conclude that fetal platelets actually do release cytokines in response to contact with collagen despite poor aggregation. Therefore, impaired aggregation to collagen cannot solely explain the minimal inflammation after fetal wounding.

In vitro analysis of fetal fibroblast collagenolytic activity.

Tissue repair in the rabbit fetus is remarkably rapid and occurs without significant inflammation or excessive collagen deposition. The objective of this study was to compare the ability of rabbit fetal and adult fibroblasts to express the matrix metalloproteinases which are thought to be critical to scar tissue remodeling. In vitro, both fetal and adult rabbit fibroblasts express procollagenase messenger RNA in a constitutive manner. Mechanical disruption of fetal fibroblast monolayers caused a twofold increase in procollagenase mRNA. In contrast, the adult rabbit fibroblast procollagenase mRNA remained unchanged. The mRNA data correlated well with enzyme protein levels. Quantitation by immunoprecipitation showed a 2.3-fold increase in fetal fibroblast procollagenase protein after mechanical injury, whereas the level in adult rabbit fibroblasts remained unchanged. However, it was noted that the constitutive levels of procollagenase mRNA and protein were higher in adult fibroblasts. Analysis of enzyme activity, by means of a fluorogenic substrate, showed that adult fibroblasts had 2.2 times more collagenase activity compared with fetal cells. After mechanical injury, the fetal fibroblast collagenase activity increased 1.3-fold compared with 1.7-fold in the adult fibroblasts. In contrast, fetal fibroblast gelatinase activity was 1.25 times greater than in adult cells and increased 1.4-fold after mechanical injury, whereas the adult profile remained unchanged. Immunolocalization studies indicated that 1 hour after mechanical injury, procollagenase was produced primarily by fibroblasts along the mechanical injury ridge. By 4 hours after injury, the ridge cells began to migrate out into the open area and procollagenase was noted in adjacent cells of both adult and fetal origin. By 12 and 18 hours, all cells throughout the monolayer were expressing procollagenase. These findings show that in vitro fetal fibroblasts actually had lower levels of procollagenase, but higher levels of gelatinase compared with adult fibroblasts. The increased gelatinase expression may explain why fetal wounds do not have excessive collagen accumulation and heal without a visible scar.

Dexamethasone abrogates the fibrogenic effect of transforming growth factor-beta in rat granuloma and granulation tissue fibroblasts.

Administration of TGF-beta, a fibrogenic inflammatory growth factor, promotes fibrosis and scarring. Dexamethasone, an anti-inflammatory steroid, inhibits wound healing and reduces fibrosis. The current studies were initiated to determine whether the co-administration of dexamethasone was able to abrogate the fibrogenic effect of TGF-beta. Polyvinyl alcohol sponges were implanted subcutaneously on the abdominal area of rats and directly injected with vehicle, dexamethasone, TGF-beta, or dexamethasone plus TGF-beta. Dexamethasone was able to block the fibrogenic effect of TGF-beta. Collagen and noncollagen protein synthesis was measured as a function of TGF-beta or dexamethasone concentrations in fibroblasts isolated from granulation tissue. Addition of dexamethasone to cultures treated simultaneously with TGF-beta blocked the fibrogenic response of TGF-beta. To study the molecular regulation of collagen gene expression by TGF-beta or dexamethasone, fibroblasts derived from granulation tissue were stably transfected with the ColCat 3.6 plasmid, which contains the rat pro alpha1(I) collagen promoter linked to the chloramphenicol acetyltransferase (CAT) gene. Dexamethasone decreased CAT activity whereas TGF-beta increased the activity of this reporter gene. The increase in CAT activity observed with TGF-beta treatment was significantly decreased when dexamethasone was added to the cultures, although CAT activity did not return to control level. Since collagen synthesis in fibroblasts treated simultaneously with dexamethasone and TGF-beta1 was found to be the same as that of untreated samples, the data indicate that there is a dexamethasone-mediated posttranscriptional regulation of pro alpha1(I) collagen mRNA. These studies demonstrate that at the in vivo level, the cellular level, and the molecular level, dexamethasone is able to block the fibrogenic effect of TGF-beta.

Cleavage of type I procollagen by human mast cell chymase initiates collagen fibril formation and generates a unique carboxyl-terminal propeptide.

The ability of human mast cell chymase and tryptase to process procollagen was examined. Purified human intestinal smooth muscle cell procollagen was incubated with human mast cell tryptase or human mast cell chymase. Purified chymase, but not tryptase, exhibited procollagen proteinase activity in the presence of EDTA. Addition of purified porcine heparin over a range of 0.1-100 microg/ml did not affect either the rate or the products of procollagen chymase cleavage. The cleavage site of chymase on the pro-alpha1(I) collagen carboxyl terminus was found to be in the propeptide region at Leu-1248-Ser-1249. Cleavage at this site suggested that the collagen products would form fibrils and confirmed the production of a unique carboxyl-terminal propeptide. Turbidometric fibril formation assay demonstrated de novo formation of chymase-generated collagen fibrils with characteristic lag, growth, and plateau phases. When observed by dark field microscopy, these fibrils were similar to fibrils formed by the action of procollagen proteinases. Thus, mast cell chymase, but not tryptase, exhibits procollagen peptidase-like activity as evidenced by its ability to process procollagen to fibril-forming collagen with concurrent formation of a unique carboxyl-terminal propeptide. These data demonstrate that mast cell chymase has a potential role in the regulation of collagen biosynthesis and in the pathogenesis of fibrosis.