| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Departments of Surgery,
*
Medicine,
and Physiology and Cellular
Biophysics,
College of Physicians &
Surgeons, Columbia University, New York, New York
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
; interleukin-6; and
matrix metalloproteinases-2, -3, and -9. In
addition, generation of thick, well-vascularized
granulation tissue was enhanced, in parallel with increased
levels of platelet-derived growth factor-B and vascular endothelial
growth factor. These findings identify a central role for RAGE in
disordered wound healing associated with diabetes, and suggest
that blockade of this receptor might represent a targeted strategy to
restore effective wound repair in this disorder.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The phases of wound healing, consisting of inflammation, proliferation, and maturation/remodeling, represent a dynamic series of events that result in transformation of an open wound, into newly formed, well-vascularized granulation tissue with overlying skin enriched in collagen and other structural elements of the extracellular matrix.4,5 In diabetes, however, abundant evidence exists that the phases of wound healing are disordered. In homeostasis, an early inflammatory phase augurs egress of neutrophils and mononuclear phagocytes from the intravascular space. These inflammatory effector cells, once released into the wounded tissue, engulf invading bacteria and promote removal of necrotic foci. In diabetes, however, decreased chemotaxis of inflammatory cells into the wound, compounded by decreased phagocytosis and intracellular killing, leads to diminished availability of factors critical for effective wound repair.2,6-8
These observations, together with those
suggesting that diabetic wound healing is associated with diminished
formation of granulation tissue and collagen, led to the premise that
impaired wound healing resulted from inadequate levels of inflammatory
cell-derived growth factors, such as platelet-derived growth factor
(PDGF) and basic fibroblast growth factor.9-13
A view has
emerged, however, that although influx of inflammatory cells into
wounded tissue is inadequate in the initial period after wounding, once
established, potent inflammatory forces unique to the diabetic
environment arise that sustain the generation of pro-inflammatory
cytokines such as tumor necrosis factor (TNF)-, and the production
of tissue destructive matrix metalloproteinases (MMPs), thereby sharply
limiting effective wound closure.14-17
It was in this context that we speculated a role for receptor for
advanced glycation end-products (RAGE) in the pathogenesis of impaired
wound healing in diabetes. RAGE, a multiligand member of the
immunoglobulin superfamily of cell surface molecules,18,19
displays enhanced expression in diabetic tissues, such as blood
vessels, atherosclerotic lesions, infected periodontal tissue, and
glomeruli.20-23
Two of the classes of RAGE’s ligands,
advanced glycation end-products (AGEs), in particular, carboxy(methyl)
lysine (CML) adducts of proteins/lipids, and EN-RAGEs, members of the
S100/calgranulin family of pro-inflammatory cytokines, modulate
pro-inflammatory responses in a receptor-dependent
manner.18,24,25
Specifically, on endothelial cells,
ligand-RAGE interaction results in induction of the procoagulant
initiator tissue factor; enhanced monolayer permeability; and
expression of adhesion molecules, such as VCAM-1, and cytokines, such
as interleukin (IL)-6.24-28
On mononuclear phagocytes
(MPs), ligation of RAGE by AGEs or EN-RAGEs modulates chemotaxis and
haptotaxis of these RAGE-bearing cells, and increases generation of
cytokines, such as TNF-, IL-6, and IL-1.24,25,29,30
In
addition, engagement of RAGE on fibroblasts results in diminished
generation of collagen.31
Taken together, these considerations led us to hypothesize that the inevitable accumulation of AGEs within diabetic skin32-34 provides a means, via RAGE, to transiently trap MPs and other inflammatory cells, thereby delaying their entry into wound foci. Once these cells gain access to the wound, however, we hypothesize that RAGE-bearing inflammatory cells, endothelia, and fibroblasts engage AGEs and EN-RAGES, thereby setting in motion a cascade of events eventuating in sustained influx, activation, and delayed egress of inflammatory effector cells from the wound. We speculate that enhanced interaction of RAGE with its pro-inflammatory ligands within the diabetic milieu provides a mechanism for sustained generation of cytokines and tissue-degradative MMPs, and diminished accumulation of collagen. In the present study, we tested the concept that blockade of RAGE in a model of impaired wound healing in genetically diabetic, db+/db+ mice might limit exaggerated pro-inflammatory responses, thereby abetting effective repair and wound closure.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Genetically diabetic, male C57BLKS/J-m+/+Leprdb mice (stock number 000642; Jackson Laboratories, Bar Harbor, ME) were used. Animals were maintained in a conventional animal facility with a 12-hour light/dark cycle. Water and standard rodent laboratory chow were provided ad libitum. At the age of 10 weeks, mice were placed in individual cages and subjected to wounding. After induction of deep anesthesia by intraperitoneal injection of ketamine (100 mg/kg; Fort Dodge, Fort Dodge, IA) and xylazine (10 mg/kg; Bayer Corporation, Shawnee Mission, KS), the hair on the surface of the mouse’s back was shaved and the skin was washed with povidone-iodine solution and alcohol. A sterilized template (1.5 cm x 1.5 cm) was placed on the midback and a full-thickness excisional wound was created by removal of the skin and panniculus carnosus. Tincture of benzoin compound (Professional Disposables, Orangeburg, NJ) was applied outside the perimeter of the wound. A semipermeable transparent dressing (Tegaderm; 3M Health Care, St. Paul, MN) was placed over the wound and sealed at the edges by benzoin. On completion of the surgical procedure, animals were injected with 1 ml of NaCl (0.9%) by intraperitoneal route. Beginning on day 3 after wounding through day 10, diabetic mice were treated with murine soluble (s) RAGE. Murine sRAGE was prepared, purified, characterized, and rendered devoid of detectable endotoxin as previously described.20 Control animals received equal amounts of vehicle, murine serum albumin (MSA) (Sigma, St. Louis, MO). Murine sRAGE or MSA was applied in sterile, endotoxin-free, phosphate-buffered saline (PBS), directly under the Tegaderm dressing; total volume, 0.1 ml/dose. In other experiments, sRAGE or MSA was administered by intraperitoneal route (50 µg/day) beginning on day 3 through 10 after wounding. After induction of deep anesthesia, blood was withdrawn from the inferior vena cava on the day of sacrifice. Levels of glycosylated hemoglobin from red blood cell lysates were determined using a kit from EG&G, Wallac, Inc. (Akron, OH). All procedures were conducted in accordance with the policies of the Institutional Animal Care and Use Committee of Columbia University.
Serial Analysis of Wound Closure
On the indicated days of, and, after wounding, the edge of the wound was traced onto a glass microscope slide. A tracing was made immediately after creation of each wound to serve as the reference point for the original area. Wound area was determined by planimetry using NIH Image 1.60 software, and reported as percent closed as calculated by the following formula: percent closure = {(area on day 0 - open area)/area on day 0} x 100. Wounds were considered closed if moist granulation tissue was no longer apparent, and the wound appeared covered with new epithelium. In all cases, to exclude pathological bacterial contamination, immediately before sacrifice, wound fluid was plated onto MacConkey II (gram-negative bacteria) and mannitol salt (gram-positive bacteria) agar plates (Becton-Dickinson and Co., Cockeysville, MD). Animals were excluded from study if any bacterial growth on these plates was observed. In our experiments, <5% of the animals were excluded from analysis because of bacterial contamination of the wound.
Tissue Collection and Preparation
At the time of sacrifice, immediately after final tracing of the
wound edges, the entire wound, including a margin of 
5 mm of
unwounded skin, was excised down to the fascia. Tissue was immediately
frozen and stored at -80°C. In other cases, tissue was placed in
buffered paraformaldehyde (4%) for fixation.
Preparation of Lysates for Immunoblotting and Zymography
Lysates were prepared from wound tissue by homogenization in Tris-buffered saline (Tris, 0.02 mol/L, pH 7.4, and NaCl, 0.150 mol/L) containing complete protease inhibitors (Boehringer Mannheim, Indianapolis, IN) using a homogenizer from Brinkmann (Westbury, NY). Lysates of wound tissue for gelatin zymography were prepared by homogenization in buffer containing Tris, 0.05 mol/L, pH 7.4; NaCl, 0.075 mol/L; and phenylmethylsulfonyl fluoride, 1 mmol/L. Protein concentrations were measured using an assay from Bio-Rad (Hercules, CA).
Immunoblotting
Protein extracts were prepared as described above and subjected to
sodium dodecyl sulfate-polyacrylamide gel electrophoresis
(Novex/Invitrogen, Carlsbad, CA). Equal amounts of protein were placed
in each lane. After electrophoresis, contents of the gels were
transferred to nitrocellulose membranes (Novex/Invitrogen). Nonspecific
binding sites on the membranes were blocked by incubation with nonfat
dry milk (5%) in Tris-buffered saline containing Tween 20 (0.1%)
(blocking buffer) for 3 hours at room temperature. Primary antibody was
incubated with each membrane for 16 hours at 4°C in blocking buffer.
After incubation, membranes were washed in Tris-buffered saline
containing Tween 20 (0.1%), followed by incubation with the
appropriate peroxidase-labeled secondary antibody (diluted in blocking
buffer) for 1 hour at room temperature. Membranes were washed in
washing buffer and sites of antibody binding were identified using a
chemiluminescence detection system (ECL; Amersham Pharmacia,
Piscataway, NJ). Molecular weights (approximate) of the bands were
identified by simultaneous electrophoresis of rainbow molecular markers
(Amersham Pharmacia). Antibodies for immunoblotting included: goat
anti-mouse IL-6 IgG (1 µg/ml; R&D Systems, Inc., Minneapolis, MN);
goat anti-mouse TNF- IgG (2 µg/ml; R&D Systems, Inc); rabbit
anti-mouse PDGF-B IgG (0.4 µg/ml; Santa Cruz Biotechnology, Santa
Cruz, CA); goat anti-mouse vascular endothelial growth factor (VEGF)
IgG (2 µg/ml; R&D Systems, Inc.); rabbit anti-mouse RAGE, prepared
and characterized as previously described20
(2 µg/ml);
mouse anti-human MMP-3 IgG (reacts with murine MMP-3) (0.1 µg/ml;
Oncogene Research Products, Cambridge, MA); mouse anti-human MMP-9 IgG
(reacts with murine MMP-9) (0.1 µg/ml; Oncogene Research Products);
and rabbit anti-S100 IgG (2.5 µg/ml; Sigma). Bands were scanned into
a laser densitometer and quantification was performed using ImageQuant
software (Molecular Dynamics, Foster City, CA). Pixel units
obtained from blots in MSA-treated mice were arbitrarily assigned a
value of 1.0. Values for bands obtained from sRAGE-treated wound
lysates were reported relative to the values obtained for MSA
treatment.
Gelatin Zymography
Wound extracts were mixed with sodium dodecyl sulfate sample buffer (Novex/Invitrogen) and electrophoresed without boiling under nondenaturing conditions and placed onto gelatin-laden gels (Novex/Invitrogen). After electrophoresis, gels were incubated in renaturing buffer for 30 minutes with gentle agitation at room temperature, and then equilibrated in developing buffer for 30 minutes. Following these procedures gels were incubated in fresh developing buffer for 16 hours with gentle agitation at 37°C. Gels were stained with GELCODE Blue Stain reagent (Pierce, Rockford, IL). Areas of gelatinase activity on the gels were evident as clear bands against a dark blue background. Bands were scanned into a laser densitometer and quantification and reporting of results was performed as above.
Histology and Immunohistochemistry
Wound tissue was fixed in buffered paraformaldehyde (4%) for 16
hours, followed by paraffin-embedding and generation of sections (5
µm thick). Certain sections were stained with hematoxylin and eosin
(H&E) or Direct Red 80 mixed with picric acid (Sigma) to yield
Picrosirius red stain for collagen. In other cases,
immunohistochemistry was performed using the following antibodies: goat
anti-mouse IgG as above (1 µg/ml); goat anti-mouse TNF- IgG (10
µg/ml; Santa Cruz Biotechnology); rabbit anti-mouse RAGE IgG (2
µg/ml); rabbit anti-EN-RAGE IgG, prepared and characterized as
previously described24
(2 µg/ml); affinity-purified
anti-CML IgG, prepared and characterized as previously
described25
(0.37 µg/ml); and rat anti-mouse Mac-3 IgG
(5 µg/ml; PharMingen, Franklin Lakes, NJ). In all cases, respective
controls with the indicated species of nonimmune IgG were used. No
specific immunostaining with control IgG was observed (not shown).
Quantification of immunohistochemistry was performed as follows:
multiple sections (at least four areas per slide, and at least 10
slides per condition) (n = 3 mice/condition)
were selected by a blinded observer and imaged using a Zeiss microscope
and an attached Sony video camera. For quantification of
immunohistochemistry on day 10, sites for study were chosen at the edge
of the wound, at the margin between normal and wounded skin. For
quantification of immunohistochemistry on days 21 and 35, sites for
study were selected at the center of the granulation tissue. For
quantification of collagen content using Picrosirius red staining,
sites were selected just below the epithelium to encompass the
granulation tissue/healing wound observed on each slide. Images were
analyzed by staining intensity on a Macintosh computer using the NIH
Image program (version 1.62) by a blinded observer.
Specifically, on each section. a density range was set to reflect the
areas stained positively with chromogen. In the case of Picrosirius red
stain, bright red-staining areas were highlighted. In all cases, the
number of pixels in the positive-staining sections was then divided by
the total number of pixels in each field. The mean ± SE is shown
for each experimental point.
Histological Score
Histological scores were determined according to previously published methods12 by one of the investigators blinded to the experimental conditions by assessment of H&E-stained sections: 1–3, none to minimal cell accumulation and granulation tissue or epithelial travel; 4–6, thin, immature granulation dominated by inflammatory cells but with few fibroblasts, capillaries or collagen deposition and minimal epithelial migration; 7–9, moderately thick granulation tissue, ranging from being dominated by inflammatory cells to more fibroblasts and collagen deposition. This score includes extensive neovascularization; epithelium can range from minimal to moderate migration; and 10–12, thick, vascular granulation tissue dominated by fibroblasts and extensive collagen deposition. Epithelium may partially to completely cover the wound.
Statistical Analysis
Data are reported as mean ± SE. Data were analyzed by analysis of variance. Where indicated, data were subjected to post hoc comparisons using a two-tailed t-test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To test the role of RAGE in diabetic wound healing, we used the
genetically diabetic db+/db+ mouse, a well-established model of insulin
resistance, hyperglycemia and impaired wound
healing.12,35,36
Wound healing in these mice, primarily
characterized by granulation tissue formation and re-epithelialization
(
60%), differs from that observed in nondiabetic heterozygous
control (m+/db+) mice, in which wound closure ensues mostly because of
contraction (90%).12
To examine comparable molecular
mechanisms of wound closure in the context of RAGE and its ligands, we
limited our studies to db+/db+ mice. RAGE blockade in diabetic mice was
achieved by administration of purified murine sRAGE, the extracellular
ligand-binding domain of RAGE that binds up ligands such as AGEs and
EN-RAGEs, thereby limiting engagement of cell surface
receptor20,21,24,27,37
. Control mice received vehicle
(MSA). Murine sRAGE or MSA, 25 µg per day, was administered topically
underneath the Tegaderm dressing from days 3 through 10 after wound
creation.
Full-thickness excisional wounds were created on the backs of diabetic
mice using a sterile template, 1.5 x 1.5 cm. Wound closure
(percent) was assessed serially in mice treated with sRAGE or MSA.
Beginning on day 7 after wounding, administration of sRAGE, 25 µg per
day, accelerated processes leading to wound closure. Mice treated with
sRAGE displayed 23.4 ± 1.5% closure compared with 18.5 ±
1.9% closure in MSA-treated animals (P = 0.02;
Figure 1a
). On day 10, compared to mice
treated with MSA, animals treated with sRAGE displayed accelerated
wound closure (19.2 ± 1.8% versus 35.5 ± 2.1%;
P = 0.01) (Figure 1a)
. Similarly, on days 14, 21, and
35 after wounding, wound closure in sRAGE-treated mice (47.9 ±
3.6%; 66.1 ± 3.9%; and 99.1 ± 0.3%, respectively) was
significantly improved compared with mice treated with MSA (34.8
± 1.9%; 48.7 ± 2.6%; and 75.2 ± 5.9%, respectively)
(Figure 1a)
. On days 21 and 35 after wounding, inspection of the wounds
created in mice treated with MSA revealed thin, sparse granulation
tissue; on day 35, wounds were not closed (Figure 1, d and f
,
respectively). In contrast, however, wounds in mice treated with sRAGE
displayed thicker granulation tissue on days 21 and 35. By day 35,
wounds were primarily healed (Figure 1, e and g)
.
|
. In contrast, mice
receiving sRAGE, 10 or 25 µg per day, displayed significantly
improved wound healing compared with mice treated with MSA on days 10,
14, and 21 (Figure 1h)
. The effects of sRAGE were not because of
restoration of normoglycemia. At sacrifice, the mean level of
glycosylated hemoglobin in mice treated with sRAGE, 8.1 ± 1.0%,
did not differ from that observed in diabetic mice treated with MSA,
7.7 ± 0.5%. In contrast, levels of glycosylated hemoglobin in
nondiabetic m+/db+ mice (n = 6) were 3.4 ±
0.2% (P < 0.00001 versus
sRAGE/MSA-treated db+/db+ mice).
Importantly, systemic administration of sRAGE, 50 µg per day, by the
intraperitoneal route, improved wound healing in a time-dependent
manner, comparable to that achieved with topically applied soluble
receptor (Figure 1i)
. By day 21 after wounding, sRAGE-treated db+/db+
mice displayed 65.7 ± 5.3% wound closure compared with 44.8
± 5.0% closure observed in MSA-treated animals
(P = 0.01). This observation suggests that
consequent to systemic administration, sRAGE in the circulation rapidly
gains access to the wound, as would be expected for a site of
dramatically increased vascular permeability. In addition, we
hypothesize that sRAGE in the bloodstream functions as a decoy
preventing pathological activation of circulating cells, such as
monocytes, that are likely to impact on reparative mechanisms operative
in the wound.
Influx of Inflammatory Cells and Formation of Granulation Tissue in db+/db+ Mice: Effect of sRAGE
These observations led us to dissect the molecular mechanisms by
which blockade of RAGE accelerated wound closure in diabetic mice.
Analysis of H&E sections scored based on extent of inflammatory cell
infiltration, neovascularization, collagen deposition, and
re-epithelialization12
provided important clues to the
mechanisms underlying the effects of sRAGE. On day 10, wounds retrieved
from mice treated topically with MSA displayed minimal influx of
inflammatory cells at the margin between unwounded and wounded skin
(Figure 2a)
. In contrast, in mice treated locally with
sRAGE, an extensive inflammatory cell infiltrate was noted at the wound
margin on day 10 (Figure 2b)
. Mean histological scores on day 10 in
wounds retrieved from MSA- versus sRAGE-treated mice
indicated that sRAGE accelerated inflammatory and reparative responses
within the diabetic wound (2.4 ± 0.3 versus 6.7
± 0.4, respectively; P = 0.00004) (Figure 2i)
. On days
14 and 21, wounds retrieved from sRAGE-treated mice displayed extensive
numbers of inflammatory cells, epithelial travel, and formation of new
blood vessels (Figure 2, d and f)
. In contrast, wounds retrieved from
MSA-treated mice continued to display fewer numbers of inflammatory
cells and thin granulation tissue at these time points (Figure 2, c and e)
. On day 14, histological scores in wounds retrieved from mice
receiving sRAGE or MSA were 7.7 ± 0.4 and 4.3 ± 0.4,
respectively; P = 0.0004 (Figure 2i)
. On day 21,
quantitative analysis revealed significantly diminished scores in mice
treated with MSA versus sRAGE, 4.7 ± 0.3
versus 9.7 ± 0.4, respectively; P =
0.00007 (Figure 2i)
. By day 35, wounds retrieved from mice treated with
MSA began to demonstrate increased numbers of inflammatory cells, along
with new blood vessel formation and collagen deposition at the
midsection. Re-epithelialization, although underway, was not nearly
complete in these mice (Figure 2g)
. In contrast, on day 35, midsections
of sRAGE-treated wounds displayed diminished numbers of inflammatory
cells, extensive re-epithelialization, neovascularization, and collagen
deposition (Figure 2h)
. Consistent with these observations,
histological scores remained decreased in MSA- versus
sRAGE-treated mice on day 35 (8.3 ± 0.4 versus
11.0 ± 0.0, respectively; P = 0.001) (Figure 2i)
.
These findings supported the hypothesis that blockade of RAGE restored
effective wound healing mechanisms by forging an appropriately limited
period of inflammation, followed by effective proliferation and repair.
|
Consistent with a role for RAGE in diabetic wound repair,
immunohistochemistry using anti-RAGE IgG revealed fewer numbers of
RAGE-bearing inflammatory cells at the wound margins on day 10 in mice
treated with MSA versus sRAGE (Figure 3, a and b)
. In
MSA-treated wounds, a very thin area of granulation tissue was formed.
However, in sRAGE-treated wounds, an extensive layer of granulation was
in place bearing increased numbers of RAGE-expressing cells
(1.5-fold greater than in MSA-treated wounds by quantitative
immunohistochemistry; P = 0.01) (Figure 3c)
. On day 21 after
wounding, in parallel with increased wound closure and histological
evidence of repair, levels of RAGE antigen were significantly decreased
in sRAGE-treated wounds compared with those receiving MSA (Figure 3, e and d
, and Figure 3f
; P = 0.00001). By day 35,
consistent with ongoing cellular activation and lack of effective
closure in MSA-treated wounds, numbers of RAGE-expressing cells were
increased 3.3-fold in the presence of MSA at the midsection,
compared with those treated with sRAGE (Figure 3, g and h
, and Figure 3i
; P = 0.00001). Because multiple studies have
suggested that expression of RAGE is enhanced at sites of ligand
deposition, we assessed levels of EN-RAGEs. EN-RAGEs, members of the
family of S100/calgranulin pro-inflammatory cytokines, are expressed in
a range of cell types, especially inflammatory cells, such as MPs,
neutrophils, and lymphocytes.24
Wound margins from
sRAGE-treated mice on day 10 displayed increased EN-RAGE antigen by
immunoblotting compared with mice treated with MSA (Figure 3j)
. As MPs
are a principal cell type within the wound that promote the
inflammatory response, we assessed levels of this key effector cell in
the wound tissue. Consistent with the concept that blockade of
RAGE accelerated the inflammatory response within the wound, increased
numbers of MPs (1.6-fold), as assessed by Mac-3 antigen, were
observed in margins of sRAGE-treated wounds on day 10 compared with MSA
(Figure 3, u and t
, and Figure 3v
; P = 0.0007). By day
21, consistent with enhanced inflammation in vehicle-treated wounds,
and decreased cellular activation in the resolving sRAGE-treated
wounds, levels of EN-RAGE antigen by immunoblotting were increased in
MSA-treated wounds compared with those wounds to which sRAGE had been
applied (Figure 3j)
.
|
1.6-fold increase in
EN-RAGE epitopes on day 21 in MSA- versus sRAGE-treated
wounds (Figure 3, n and o
, and Figure 3p
; P = 0.00001).
By day 35 after wounding, levels of EN-RAGE antigen by immunoblotting,
were increased 1.3-fold in mice treated with MSA versus
sRAGE (Figure 3j)
. Immunohistochemical studies confirmed these
observations. On day 35, wounds from MSA-treated mice displayed an
1.9- and 2.5-fold increase in EN-RAGE and Mac-3 antigens in wound
midsections (Figure 3, q and s, and z
and bb, respectively), compared
with sRAGE-treated wounds (Figure 3, r and s
, and aa and bb,
respectively).
As studies have indicated that levels of AGE epitopes are increased in
diabetic skin,32-34
and that administration of sRAGE
effects reduced levels of AGE epitopes in vivo in murine
models,20
we assessed levels of AGE/CML epitopes in wound
tissue of db+/db+ mice by immunohistochemistry using affinity-purified
anti-CML IgG.25
On day 21 after wound creation, levels of
CML epitopes in sRAGE-treated wounds were reduced 1.4-fold compared
with wounds receiving MSA (Figure 3
, dd and cc, and Figure 3e
e;
P = 0.05). By day 35, levels of CML epitopes were
2.5-fold greater in MSA-treated wounds, within the granulation
tissue, blood vessels, and cellular elements at the wound midsection,
compared with wounds treated with sRAGE (Figure 3
, ff and gg, and
Figure 3h
h; P = 0.002). In sRAGE-treated wounds, newly
formed epithelium, cellular elements, blood vessel structures, and
collagenous deposits were primarily free of CML epitopes (Figure 3g
g).
Expression of Cytokines and MMPs in Wound Tissues: Effect of sRAGE
Taken together, these observations suggested that although influx
of inflammatory cells into vehicle-treated diabetic wounds was delayed,
once in place, sustained cellular activation followed, leading to
failure of remodeling and wound closure by day 35. Indeed, by day 21
after wounding, the tide appearing already turning in sRAGE-treated
wounds, in that levels of RAGE and its ligands were already receding.
To test the effects of blocking ligand-RAGE axis on levels of
tissue-destructive molecules within the wound, we assessed levels of
two potent pro-inflammatory cytokines. On day 10, there was no
difference in expression of TNF- in wounds treated with sRAGE or MSA
(Figure 4, a–d)
. However, on days 21 and
35, levels of TNF- antigen were increased 1.8- and 1.4-fold by
immunoblotting in mice treated with MSA versus sRAGE,
respectively (Figure 4a)
. Levels of TNF--expressing cells were
increased 1.6-fold in MSA-treated wounds compared with those wounds
receiving sRAGE on day 21 (Figure 4, e and f
, and Figure 4g
;
P = 0.000001). Further, by day 35 after wounding, an
1.6-fold increase in levels of TNF- was observed in MSA-
versus sRAGE-treated wounds at the midsection (Figure 4, h and i
, and Figure 4j
; P = 0.001). Additional support
for this concept was evident on assessment of levels of IL-6 in
diabetic wounds. Although there was no appreciable difference in levels
of IL-6 at 10 days after wounding in MSA- versus
sRAGE-treated mice (Figure 4, k–n)
, on days 21 and 35, immunoblotting
revealed increased expression of IL-6 antigen in MSA-treated wounds
versus those diabetic wounds treated with sRAGE (1.3- and
1.4-fold, respectively) (Figure 4k)
. Similarly, IL-6 antigen-expressing
cells in the midsections of wounds on day 21 were increased 1.5-fold
in MSA-treated wounds versus those treated with sRAGE
(Figure 4, o and p
, and Figure 4q
; P = 0.000001).
Similarly, by day 35 after wounding, levels of IL-6 were increased
1.9-fold in mice treated with MSA versus sRAGE (Figure 4, r and s
, and Figure 4t
; P = 0.0002).
|
in the inflammatory response is the generation of activated
forms of MMPs.38
Once activated, MMPs mediate destruction
of newly formed collagen, as well as other structural matrix elements
and molecules within the wound milieu. Immunoblotting studies revealed
that levels of MMP-3 antigen were reduced 1.3-fold in sRAGE-treated
versus MSA-treated wounds on day 21 (Figure 5a)
, and, by day 35, levels of MMP-3
antigen were increased 5.0-fold in wounds retrieved from mice
treated with MSA versus sRAGE (Figure 5a)
. In addition,
levels of MMP-9 antigen, nearly identical on day 10 after wounding in
both groups, were elevated 1.5-fold and 2.0-fold in MSA-treated
wounds versus those receiving sRAGE on days 21 and 35 after
wound creation, respectively (Figure 5b)
. Analogous to these findings
at the antigen level, zymography revealed an 1.4-fold and
1.9-fold increase in active MMP-9 on days 21 and 35 in MSA-
versus sRAGE-treated wounds, respectively (Figure 5c)
.
Furthermore, levels of MMP-2 activity were enhanced in MSA-treated
wounds on day 35 (Figure 5d)
. Consistent with the concept that
diminished MMP antigen and activity were evident at the later stages of
wound repair in mice treated with sRAGE, Picrosirius red staining
revealed significantly increased collagen in wounds treated with sRAGE
compared with MSA on days 21 (Figure 5, f and e
, and Figure 5g
;
P = 0.008) and 35 (Figure 5, i and h
, and Figure 5j
;
P = 0.02).
|
An essential role of inflammatory cells infiltrating wounded
skin is to promote generation of growth factors. These factors,
critical for a wide array of functions designed to limit inflammation,
enhance neovascularization, and modulate the laying down of new
collagen, are diminished in diabetic wounds, at least in part, because
of diminished influx of inflammatory effector cells in the first days
after wounding.12
Consistent with the premise that
blockade of RAGE accelerated beneficial inflammatory responses in
diabetic wounds, immunoblotting of lysates prepared from sRAGE-treated
wounds on day 10 revealed an 1.5-fold increase in PDGF-B epitopes
compared with wounds treated with MSA (Figure 6a)
. In addition, epitopes for VEGF in
lysates prepared from sRAGE-treated wounds on days 10 and 21 were
enhanced 2.2-fold, and 1.5-fold, respectively, by
immunoblotting, compared with wounds treated with MSA (Figure 6b)
.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and proteases in the wound were identified as key
biochemical/molecular factors predictive of failure of closure in human
wounds.41
In addition to exaggerated generation of
pro-inflammatory factors, recent observations suggested that blunted
apoptosis of inflammatory cells within the wound contributes to,
en balance, sustained pro-inflammatory forces within the
diabetic wound.42,43 It is thus not surprising that recent strategies aimed at introducing pharmacological levels of growth factors such as PDGF-BB and basic fibroblast growth factor into diabetic wounds have been met with equivocal success.44-46 In this context, the molecular targets of MMPs are myriad, and include not only elements of the extracellular matrix, but also growth factors and other polypeptides within the wound milieu. Thus, the observation that human and murine diabetic chronic wounds are enriched in MMPs provides support for the premise that impaired growth factor availability may limit effectiveness of this strategy. Further, the observation that chronic diabetic wounds are best characterized by a sustained inflammatory response14-17 suggests that exogenous application of single or combinations of growth factors may not address the fundamental impairment in the diabetes.
Taken together, these considerations underscore the concept that in
wound healing, a fine balance must be struck between appropriately
limited inflammatory responses, and those that, when sustained, portend
ongoing tissue degradation. Our findings place RAGE at the center of a
cascade of events that disturbs the equilibrium between beneficial and
injurious inflammation in the diabetic wound. We speculate that chronic
accumulation of AGEs within the skin and subcutaneous elements serves
as a temporary glue, delaying the entry of blood- and tissue-derived
inflammatory cells such as MPs into the wound site. Once in the wound,
however, RAGE-bearing MPs and other cellular effector cells, such as
lymphocytes, endothelial cells, and fibroblasts, interact with AGEs and
EN-RAGEs (the latter released from invading inflammatory cells) in a
sustained manner. These events prime a spiraling cascade of cellular
activation, and a vicious cycle of cytokine and MMP generation,
mediated, at least in part, by delayed egress of activated inflammatory
cells from the wound milieu (Figure 7)
.
Blockade of RAGE restores physiological migration of inflammatory cells
into, and then out of, inflamed foci, as well as their limited
activation, thereby re-setting molecular cues within the wound leading
to effective inflammation and wound repair. Indeed, on day 10 after
wounding, administration of sRAGE accelerated the inflammatory
response, as indicated by increased levels of RAGE and EN-RAGEs,
macrophages, and pro-inflammatory mediators compared with
vehicle-treated wounds. However, by days 21 and 35, in parallel with
increasing wound closure and histological evidence of thick granulation
tissue and re-epithelialization, levels of RAGE, its ligands,
macrophages, and pro-inflammatory/tissue-degradative molecules were
reduced in wounds treated with sRAGE compared with those to which MSA
had been applied.
|
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by grants from the Surgical Research Fund, Department of Surgery, College of Physicians and Surgeons, Columbia University; the United States Public Health Service grant HL60901 (to A. M. S., S. F. Y., and D. M. S.); grant JDRF4-200-945 from the Juvenile Diabetes Research Foundation International (to A. M. S. and D. M. S.); and BWC APP 2601 from the Burroughs Wellcome Fund. L. B. is a Postdoctoral Research Fellow of the Juvenile Diabetes Research Foundation International (grant JDRF 3-2000-112), and A. M. S. is a recipient of a Burroughs Wellcome Fund Clinical Scientist Award in Translational Research.
Accepted for publication April 13, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(carboxymethyl)lysine modifications of proteins are ligands for RAGE that activate cell signalling pathways and modulate gene expression. J Biol Chem 1999, 274:31740-31749This article has been cited by other articles:
 |
|
 |
|
  G. W. Martens, M. C. Arikan, J. Lee, F. Ren, D. Greiner, and H. Kornfeld Tuberculosis Susceptibility of Diabetic Mice Am. J. Respir. Cell Mol. Biol., November 1, 2007; 37(5): 518 - 524. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Marsche, B. Weigle, W. Sattler, and E. Malle Soluble RAGE blocks scavenger receptor CD36-mediated uptake of hypochlorite-modified low-density lipoprotein FASEB J, October 1, 2007; 21(12): 3075 - 3082. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Herold, B. Moser, Y. Chen, S. Zeng, S. F. Yan, R. Ramasamy, J. Emond, R. Clynes, and A. M. Schmidt Receptor for advanced glycation end products (RAGE) in a dash to the rescue: inflammatory signals gone awry in the primal response to stress J. Leukoc. Biol., August 1, 2007; 82(2): 204 - 212. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Ding, A. Kantarci, J. A. Badwey, H. Hasturk, A. Malabanan, and T. E. Van Dyke Phosphorylation of Pleckstrin Increases Proinflammatory Cytokine Secretion by Mononuclear Phagocytes in Diabetes Mellitus J. Immunol., July 1, 2007; 179(1): 647 - 654. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Campo, P. Morbini, M. Zorzetto, C. Tinelli, E. Brunetta, C. Villa, C. Bombieri, M. Cuccia, C. Agostini, V. Bozzi, et al. Expression of Receptor for Advanced Glycation End Products in Sarcoid Granulomas Am. J. Respir. Crit. Care Med., March 1, 2007; 175(5): 498 - 506. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Ding, A. Kantarci, H. Hasturk, P. C. Trackman, A. Malabanan, and T. E. Van Dyke Activation of RAGE induces elevated O2- generation by mononuclear phagocytes in diabetes J. Leukoc. Biol., February 1, 2007; 81(2): 520 - 527. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Alikhani, C. M. MacLellan, M. Raptis, S. Vora, P. C. Trackman, and D. T. Graves Advanced glycation end products induce apoptosis in fibroblasts through activation of ROS, MAP kinases, and the FOXO1 transcription factor Am J Physiol Cell Physiol, February 1, 2007; 292(2): C850 - C856. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. G. Raman, P. L. Sappington, R. Yang, R. M. Levy, J. M. Prince, S. Liu, S. K. Watkins, A. M. Schmidt, T. R. Billiar, and M. P. Fink The role of RAGE in the pathogenesis of intestinal barrier dysfunction after hemorrhagic shock Am J Physiol Gastrointest Liver Physiol, October 1, 2006; 291(4): G556 - G565. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Cheng, A. S. Pollock, R. Mahimkar, J. L. Olson, and D. H. Lovett Matrix metalloproteinase 2 and basement membrane integrity: a unifying mechanism for progressive renal injury FASEB J, September 1, 2006; 20(11): 1898 - 1900. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Goldin, J. A. Beckman, A. M. Schmidt, and M. A. Creager Advanced Glycation End Products: Sparking the Development of Diabetic Vascular Injury Circulation, August 8, 2006; 114(6): 597 - 605. [Abstract] [Full Text] [PDF]
|
|
green
segments), however, inflammatory cells such as
MPs, along with other RAGE-bearing effector cells
(such as fibroblasts and endothelial
cells), interact with AGEs and EN-RAGEs.
Activation of RAGE initiates a cascade of cellular activation, typified
by generation of cytokines and MMPs (green
arrows). Delayed egress of inflammatory
cells portends ongoing stimulation of RAGE, sustained inflammation, and
failure of reparative/remodeling processes. We speculate that blockade
of RAGE both hastens and appropriately limits the inflammatory phase of
wound healing in diabetes, thereby promoting closure.