| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the School of Biological Sciences,*
University of
Manchester, Manchester, United Kingdom, and the Laboratories of Cell
Regulation and Carcinogenesis
Experimental Carcinogenesis,
National Cancer
Institute, Bethesda, Maryland
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
TGF-ß1 has also been implicated in various fibrotic disorders, such as glomerulonephritis5 and pulmonary fibrosis.6 Increased levels of plasma TGF-ß have been found to correlate with increased fibrogenesis after bone marrow transplantation therapy in patients with advanced breast cancer.7 Intravenous administration of recombinant TGF-ß1 to rats induces fibrotic lesions in the liver, kidneys, pancreas, and testes,8 suggesting an endocrine-like effect of TGF-ß1.
We have used the recently developed transgenic mouse lines that express high levels of active TGF-ß19 to investigate the role of elevated systemic levels of active TGF-ß1 on wound healing. The liver fibrosis and delayed liver regeneration after partial hepatectomy characteristic of these transgenic lines has been shown to result directly from the overexpression of TGF-ß1,10 and in line 25 mice, the characteristic kidney fibrosis and kidney failure has also been shown to be due to the high circulating levels of TGF-ß1 driven by the transgene.11 Based on these observations, we wished to test the hypothesis that elevated plasma TGF-ß1 would enhance scarring in cutaneous wounds. As polyvinyl alcohol (PVA) sponges have frequently been used to assess wound healing, we evaluated healing in both incisional and PVA sponges. Surprisingly, cutaneous wounds in transgenic, TGF-ß1-overexpressing mice healed with less scarring than control mice, whereas the sponges showed the opposite response, with the transgenic mice demonstrating an enhanced rate of cellular influx and matrix deposition into the sponges compared with controls.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Experimental Model
Animals were anesthetized by intraperitoneal injections of
avertin. All animals were males between 4 and 6 weeks of age. Four
0.75-cm full-thickness incisional wounds were placed on the dorsal
skin, equidistant from the midline and adjacent to the four limbs.
These wounds were left to heal by secondary intention (Figure 1a)
. Two PVA sponges (0.5 cm x 0.5
cm and weighing 8 mg each) were inserted into subcutaneous pockets on
the ventral wall of the abdomen through two ventral incisions that were
subsequently sutured with 5.0 ethilon (Ethicon) to prevent
extrusion of the sponges (Figure 1b)
. Animals were recovered in their
cages and were fed mouse chow and water ad libitum. In
total, 29 control mice and 26 mice from line 25 were wounded. Four mice
from line 25 died within 2 weeks of surgery. The remaining animals were
killed on days 7, 14, and 80 after wounding by carbon dioxide narcosis,
and the cutaneous wounds and sponges were harvested. From each animal,
the dorsal cutaneous wounds and sponges were processed for routine
histology, immunocytochemistry, and hydroxyproline assay as described
previously.12,13
The sponges were carefully dissected free
of the surrounding capsule before processing. All animal protocols were
approved by institutional (National Institutes of Health) review.
|
The wounds and sponges were fixed in 10% formal saline, processed for paraffin embedding, sectioned at 7-mm thickness, and stained with picrosirius red to enhance polarization of collagen fibers.14 The differences in the architecture of the neodermis of wounds in the control and transgenic mice were assessed using a polarizing microscope (Zeiss, Oberkochen, Germany) as described previously.13
Histological sections of the PVA sponges were also stained with hematoxylin and eosin to assess the cellularity and presence of granulation tissue within the sponges. A graticule (Leitz) was used in the eye-piece of the microscope, and cells present within two randomly selected defined areas (0.05 mm2) at 40x magnification were counted. The cellularity of three sections from each PVA sponge was quantified. Note the randomly defined areas did not include the capsules surrounding the sponges but the spaces within the sponges.
Immunocytochemistry
TGF-ß isoforms and their receptors were immunodetected in paraffin-embedded tissue sections using the immunoperoxidase staining technique. Briefly, sections were dewaxed and treated with 0.3% Triton X-100 in Tris-buffered saline (TBS) for 15 minutes. Endogenous peroxidase was quenched with 0.6% hydrogen peroxide in methanol for 1 hour at room temperature, followed by digestion with sheep testicular hyaluronidase (Sigma Chemical Co., Poole, UK) at a concentration of 1 mg/ml in sodium acetate buffer, pH 5.5, with 0.85% sodium chloride for 0.5 hour at 37°C. Normal goat serum (Vector Laboratories, Burlingame, CA) was used to block nonspecific antibody binding. Sections were then incubated with the primary antibody diluted in the blocking serum overnight at 4°C. After thorough washing in TBS/0.1% bovine serum albumin, the sections were incubated with the secondary antibody for 1 hour at room temperature followed by incubation with the ABC complex for 1 hour at room temperature. Sections were thoroughly washed after each step except after blocking with normal goat serum. Sections were then incubated with 0.05% diaminobenzidine (DAB) until the brown substrate was formed, rinsed in distilled water, counterstained with Harris's hematoxylin, rinsed in water, dehydrated, and mounted in Practamount. The Vectastain elite ABC kit (Vector Laboratories) was used for this protocol. For all primary antibodies, biotinylated goat anti-rabbit IgG from the Vectastain ABC kit was used at a dilution of 1/200 as the secondary antibody. Rabbit polyclonal antibodies to TGF-ßs15 and their receptors (Santa Cruz Biotechnologies, Santa Cruz, CA) were used as follows: TGF-ß1, anti-P(1–30) LC, at 7 mg/ml; TGF-ß2, anti-P(50–75), at 1.5 mg/ml; TGF-ß3, anti-P(50–60), at 2.4 mg/ml; TGF-ß type I receptor, anti-P(482–501) RI (R-20), at 1:100 dilution; TGF-ß type II receptor, anti-P(550–565) RII (C-16), at 1:100 dilution.
As controls for the staining procedure, sections were incubated with an
equivalent concentration of irrelevant rabbit IgG (Serotec, Oxford, UK)
instead of the primary antibodies, and sections were incubated with
normal goat serum (omitting the primary antibody) with the rest of the
protocol unchanged. Nonspecific brown cellular staining was not
observed in any of the sections used as controls for the staining
procedure. The specificity of the antibodies used has been established
previously by Flanders et al15
by using preincubation of
the antibodies with the relevant peptides. For each primary antibody,
three sections each from at least four wounds from each group of
animals at each of days 7 and 14 were analyzed for immunoreactivity.
Figures 4 to 8
are representative of the results obtained.
|
|
|
All materials were of analytical grade and purchased from BDH Chemicals (Poole, UK) or from Sigma Chemical Co. unless otherwise stated.
The cutaneous wounds were microdissected flush with the wound margin, freeze dried, weighed, and stored at -70°C until further use. The sponges were carefully dissected free from the surrounding capsule, freeze dried, weighed, and stored at -70°C until further use. Normal, unwounded skin and unused sponges were also processed similarly. Hydroxyproline content of the samples was determined using the method of Stegmann and Stalder16 and expressed as µg/mg dry weight of samples.
Statistical Analyses
The hydroxyproline contents of sponges and wounds were compared for the wild-type and transgenic mice at three different time points. The distribution of the data was tested, and some degree of skewness was detected in most cases. Log transformation was found to be the most appropriate and consistent normalizing transformation across all three time points for the sponges and cutaneous wounds, and t-tests were then performed to compare differences between the two groups. Results are expressed as the geometric means with their 95% confidence intervals as well as the t statistic and P values. A P value of less than 0.05 indicates a significant difference between groups. Similar statistical analyses were performed for the cell counts in the PVA sponges.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cutaneous Scarring: Macroscopic Appearance
Surprisingly, the cutaneous wounds in transgenic mice healed with
a fine linear scar that was barely discernible 80 days after wounding
(Figure 2b)
. By contrast, wounds in the
control mice healed with firm, raised, obvious scars (Figure 2a)
. These
differences in the macroscopic appearances of scars were apparent as
early as 14 days after wounding.
|
The histological appearance of wounds was consistent with the
macroscopic appearance of the healed wounds. Although wounds in the
control mice generally appeared wider than wounds in the transgenic
mice, all wounds had re-epithelialized by 7 days after wounding. As
expected, sections of wounds from control mice stained with picrosirius
red and visualized through polarized light demonstrated characteristics
of scarring, including obvious compact, parallel arrangement of
collagen fibers in the neodermis of control wounds (Figure 3a)
rather than the basket-weave
appearance of normal dermis (Figure 3b)
. By contrast, the neodermis of
wounds in transgenic mice (Figure 3c)
showed some degree of
basket-weave organization of collagen and more closely resembled the
normal dermis (Figure 3d)
, reflecting a better quality of healing with
reduced scarring.
|
.
|
The hydroxyproline content of cutaneous wounds from transgenic
mice harvested 7 days after wounding was, on average, less than half
the hydroxyproline content of wounds from control mice. There were no
significant differences in the hydroxyproline content of wounds from
the wild-type and transgenic mice harvested at later time points (Table 2)
. However, it was difficult to dissect
the wounds from the surrounding normal skin of transgenic mice at later
times due to the better quality of healing.
|
.
|
The three isoforms of TGF-ß have previously been shown to be expressed in unique patterns in epidermis and to be regulated independently.17 To understand the effects of the systemic overexpression of TGF-ß1 driven by the albumin promoter, which is expressed most strongly in the liver,9 on the expression of the TGF-ß isoforms and their receptors in the skin of the transgenic mice, we assessed their patterns of immunohistochemical staining at both 7 and 14 days after wounding.
Surprisingly, the epidermis and neodermis of cutaneous wounds from
transgenic mice (Figure 4b)
stained much
less intensely for TGF-ß1 than wounds from control mice (Figure 4a)
.
By contrast, there was a marginal increase in the staining of TGF-ß2
in cutaneous wounds, principally in the epidermis of the transgenic
mice (Figure 5b)
compared with control
mice (Figure 5a)
. Most notably, there was a marked increase in
immunostaining for TGF-ß3 in the cutaneous wounds from transgenic
mice (Figure 6b)
compared with cutaneous
wounds from control mice (Figure 6a)
, specifically in the cellular
staining and depth of neodermal and subcutaneous staining.
The staining pattern observed in the PVA sponges differed from that
observed in the cutaneous wounds, in that PVA sponges from the
transgenic mice (Figures 4d, 5d, and 6d)
stained markedly more
intensely for all three TGF-ß isoforms than the PVA sponges from
control mice (Figures 4c, 5c, and 6c)
.
Whereas cutaneous wounds from control and transgenic mice did not
differ in immunoreactivity for TGF-ß receptor I (Figure 7, a and b)
, the immunoreactivity for the
type II receptor was elevated in wounds from the transgenic mice
(Figure 8b)
compared with controls
(Figure 8a)
. Similar to that observed for the TGF-ß isoforms,
immunostaining for both RI and RII was more marked in PVA sponges
from transgenic mice (Figures 7d and 8d)
compared with those from
control mice (Figures 7c and 8c)
.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Increased tissue expression of TGF-ß1 has been implicated in various fibrotic disorders,5,6 and elevated plasma levels of TGF-ß1 have also been shown to result in fibrotic sequelae in tissues such as liver, lung, and kidney.7,11 Moreover, numerous studies have shown either systemic18 or topical TGF-ß1 both to accelerate the rate of healing and to enhance scarring in cutaneous wounds3,4,19 as well as in other injury models.20 By contrast, we have shown that enhanced expression levels of TGF-ß3 relative to TGF-ß1 in wounds, achieved either by direct application of TGF-ß3 or by suppression of TGF-ß1 activity by use of isoform-specific blocking antibodies, reduces the scarring and results in a better dermal architecture.4 TGF-ß3 is thought to exert its anti-scarring activity in part by antagonizing TGF-ß1.21 In our studies presented here, cutaneous wounds from transgenic mice overexpressing TGF-ß1 showed reduced scarring accompanied by an increase in immunostaining for TGF-ß3 and a decrease in the immunostaining for TGF-ß1 compared with wounds from control mice, reinforcing our previous observations that the relative ratio of the three TGF-ß isoforms is critical for control of cutaneous scarring.4
The mechanisms whereby elevated circulating levels of active TGF-ß1 in these transgenic mice lead to lower immunoreactivity for TGF-ß1 and increased levels of TGF-ß3 in cutaneous wounds compared with control mice are not known. Auto-induction and cross-regulation of the three TGF-ß isoforms is complex and tissue specific.22,23 It may be that the high levels of systemic TGF-ß1 down-regulate subsequent expression of TGF-ß1 by desensitization of the TGF-ß receptor and messenger signaling pathways. Recent data characterizing members of the Smad family of signaling proteins, which mediate signals from the serine-threonine kinase receptors of the TGF-ß superfamily,24 show that two of these proteins, Smad 6 and Smad 7, which inhibit signal transduction from TGF-ß receptors, are up-regulated strongly by TGF-ß, suggesting that they can function to truncate the downstream signal in the presence of excessive ligand.24,25 Such an inhibitory mechanism could potentially impact most severely on expression of TGF-ß1, which is strongly autoregulated by AP-1 sites in its promoter.26
TGF-ß signals through a heteromeric complex of the type I and II TGF-ß receptors.24,27 We investigated the TGF-ß receptor profiles of cutaneous wounds and found little difference in the immunoreactivity for TGF-ß receptor I between transgenic and control mice. By contrast, cutaneous wounds from transgenic mice were more immunoreactive for TGF-ß receptor II than wounds from control mice. How changes in the relative ratios of these two receptors may affect downstream targets is poorly understood. However, several lines of investigation suggest that a preponderance of type II receptors favors antiproliferative responses, whereas a preponderance of type I receptors is profibrotic.28-30 These observations are consistent with our observations of reduced scarring and collagen synthesis in cutaneous wounds of the transgenic mice where the relative ratios of the receptors are shifted in favor of the type II receptor.
Importantly, the results of this investigation also demonstrate that host responses to subcutaneously implanted PVA sponges are different from the response to cutaneous wounding. In particular, the patterns of local induction of TGF-ß receptors and cross-regulation of TGF-ß isoforms observed in the cutaneous wounds are altered in the PVA sponge. This has important implications as PVA sponges are often used as models of the wound-healing response,31 whereas they clearly more accurately reflect the host response to a relatively inert foreign body. At 7 and 14 days after wounding, the PVA sponges from transgenic mice had more cells and a higher hydroxyproline content than sponges from control mice. It is possible that elevated levels of plasma active TGF-ß1 may lead to prestimulation of monocytes causing the rapid infiltration of cells and increase in collagen deposition seen in the PVA sponges implanted in the transgenic mice. This is in keeping with reported data where exogenous addition of TGF-ß1 increased the cellularity and the hydroxyproline/protein content of subcutaneously implanted wound chambers/sponges in rodents.31,32 By contrast, the hydroxyproline content of cutaneous wounds from transgenic mice was lower than that from control mice. It may be that the PVA sponges, which are bathed in and sequester plasma, reflect the fibrogenic response of high levels of plasma active TGF-ß1 as seen in various organs, such as the liver and the kidneys.9,11 Differences in organ perfusion or possibly in the ability to recognize and take up distinct molecular complexes of circulating TGF-ß1 may influence the response of different tissues to high circulating levels of TGF-ß1.33
Unlike cutaneous wounds, the immunoreactivity for TGF-ß1 in the PVA sponges from transgenic mice was higher than in the sponges from control mice. The PVA sponges from transgenic mice also stained more intensely for TGF-ß2 and TGF-ß3 as well as for TGF-ß receptors I and II. Recent in vitro data showed the ability of TGF-ß3 to regulate collagen synthesis by dermal fibroblasts through TGF-ß1-dependent and -independent pathways. Exogenous addition of small amounts of TGF-ß3 when added in combination with TGF-ß1 stimulated procollagen and TGF-ß1 mRNA whereas greater amounts of TGF-ß3 down-regulated procollagen and TGF-ß1 mRNA.21 It would appear that the increased fibrogenic response seen in the PVA sponges from transgenic mice is a function both of the elevated levels of the three TGF-ß isoforms and their receptors and the different ratios of both ligands and receptors, compared with that in cutaneous wounds of the same mice. Finally, these data further demonstrate the complex regulation of TGF-ßs in vivo and tend to negate our simple hypothesis to explain individual variation in the scarring response in man, ie, that people who scar badly have higher systemic levels of circulating TGF-ß1.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
M. Shah was supported by a travel grant from the Wellcome Trust.
Accepted for publication January 13, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  F. Meng, X. Cheng, L. Yang, N. Hou, X. Yang, and A. Meng Accelerated re-epithelialization in Dpr2-deficient mice is associated with enhanced response to TGF{beta} signaling J. Cell Sci., September 1, 2008; 121(17): 2904 - 2912. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Mori, K. T. Power, C. M. Wang, P. Martin, and D. L. Becker Acute downregulation of connexin43 at wound sites leads to a reduced inflammatory response, enhanced keratinocyte proliferation and wound fibroblast migration J. Cell Sci., December 15, 2006; 119(24): 5193 - 5203. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Bandyopadhyay, J. Fan, S. Guan, Y. Li, M. Chen, D. T. Woodley, and W. Li A "traffic control" role for TGF{beta}3: orchestrating dermal and epidermal cell motility during wound healing J. Cell Biol., March 27, 2006; 172(7): 1093 - 1105. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.E. van Beurden, J.W. Von den Hoff, R. Torensma, J.C. Maltha, and A.M. Kuijpers-Jagtman Myofibroblasts in Palatal Wound Healing: Prospects for the Reduction of Wound Contraction after Cleft Palate Repair J. Dent. Res., October 1, 2005; 84(10): 871 - 880. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Dai, F. Losy, A.-M. Guinault, C. Pages, I. Anegon, P. Desgranges, J.-P. Becquemin, and E. Allaire Overexpression of Transforming Growth Factor-{beta}1 Stabilizes Already-Formed Aortic Aneurysms: A First Approach to Induction of Functional Healing by Endovascular Gene Therapy Circulation, August 16, 2005; 112(7): 1008 - 1015. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Komi-Kuramochi, M. Kawano, Y. Oda, M. Asada, M. Suzuki, J. Oki, and T. Imamura Expression of fibroblast growth factors and their receptors during full-thickness skin wound healing in young and aged mice J. Endocrinol., August 1, 2005; 186(2): 273 - 289. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. W. Chu, J. G. Rino, R. B. Wexler, K. Campbell, R. J. Harbeck, and R. J. Martin Mycoplasma pneumoniae infection increases airway collagen deposition in a murine model of allergic airway inflammation Am J Physiol Lung Cell Mol Physiol, July 1, 2005; 289(1): L125 - L133. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Mazzieri, V. Jurukovski, H. Obata, J. Sung, A. Platt, E. Annes, N. Karaman-Jurukovska, P.-E. Gleizes, and D. B. Rifkin Expression of truncated latent TGF-{beta}-binding protein modulates TGF-{beta} signaling J. Cell Sci., May 15, 2005; 118(10): 2177 - 2187. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. L. Watts, E. M. Sampson, G. S. Schultz, and M. A. Spiteri Simvastatin Inhibits Growth Factor Expression and Modulates Profibrogenic Markers in Lung Fibroblasts Am. J. Respir. Cell Mol. Biol., April 1, 2005; 32(4): 290 - 300. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. WERNER and R. GROSE Regulation of Wound Healing by Growth Factors and Cytokines Physiol Rev, July 1, 2003; 83(3): 835 - 870. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Kim, X. Liu, T. Kobayashi, T. Kohyama, F.-Q. Wen, D. J. Romberger, H. Conner, P. S. Gilmour, K. Donaldson, W. MacNee, et al. Ultrafine Carbon Black Particles Inhibit Human Lung Fibroblast-Mediated Collagen Gel Contraction Am. J. Respir. Cell Mol. Biol., January 1, 2003; 28(1): 111 - 121. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q H Song, V E Klepeis, M A Nugent, and V Trinkaus-Randall TGF-{beta}1 regulates TGF-{beta}1 and FGF-2 mRNA expression during fibroblast wound healing Mol. Pathol., June 1, 2002; 55(3): 164 - 176. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. G. Lee, R. J. Homer, Z. Zhu, S. Lanone, X. Wang, V. Koteliansky, J. M. Shipley, P. Gotwals, P. Noble, Q. Chen, et al. Interleukin-13 Induces Tissue Fibrosis by Selectively Stimulating and Activating Transforming Growth Factor {beta}1 J. Exp. Med., September 17, 2001; 194(6): 809 - 822. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Chegini, K. Kotseos, Y. Zhao, B. Bennett, F. W. McLean, M. P. Diamond, L. Holmdahl, and J. Burns Differential expression of TGF-{beta}1 and TGF-{beta}3 in serosal tissues of human intraperitoneal organs and peritoneal adhesions Hum. Reprod., June 1, 2001; 16(6): 1291 - 1300. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
