University of Southern California Dissertations and Theses

Regulation of matrix metalloproteinase-9 in cutaneous wound healing

(USC Thesis Other)

Regulation of matrix metalloproteinase-9 in cutaneous wound healing

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

REGULATION OF MATRIX METALLOPROTEINASE-9 IN CUTANEOUS
WOUND HEALING

by

Wesley Andrew Grimm

__________________________________________________________________

A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(EXPERIMENTAL AND MOLECULAR PATHOLOGY)

December 2008

Copyright 2008 Wesley Andrew
Grimm

ii

Acknowledgments

My warmest gratitude goes to: My mentor, Dr. Yuan-Ping Han, for his patience
and enthusiasm for learning and teaching. My other mentors, Dr. Warren Garner and Dr.
Cheng-Ming Chuong for their guidance and understanding. My lab-mates Chunli, Lan,
Yujiro, and Lily for their support. Dr. Tim Gallaher of the USC proteomics core and Dr.
Klaus Linse of the University of Texas proteomics core facility for their insight and
advice, and the department of Animal Resources for their care and support.

iii

Table of Contents

Acknowledgements ii

List of Figures iv

Abstract v

Introduction 1

Materials and Methods 3

Results 7

Discussion 23

References 26

iv

List of Figures

Figure 1: Skin architecture is highly variable 10

Figure 2: Cutaneous injury is associated with inflammatory 12
cytokines and MMP-9

Figure 3: Animal models of cutaneous injury and 13
wound healing recapitulate some characteristics of human
wound healing

Figure 4: During wound healing, MMP-9 is controlled by 16
inflammatory cytokines and growth factors

Figure 5: Epidermal keratinocytes and dermal fibroblasts 18
provide different cytokines and MMPs to the wound
environment in response to TNF-alpha

Figure 6: MMP-9 is activated by multiple mechanisms 20

Figure 7: Skin extract contains the activating enzyme and 22
can be used to study the mechanism of MMP-9 activation in vitro

v

Abstract

BACKGROUND: Matrix metalloproteinases (MMPs) represent a family of
zinc-dependent proteases that contain a conserved pro-peptide domain. MMP-9 is a
gelatinase known for degrading type IV collagen among its potential substrates. MMP-9
is implicated in a myriad of different biological processes and pathologies such as wound
healing, and cancer. Thus, strict transcriptional and post-translational regulation of MMP-
9 is important. Latent pro-MMPs have a conserved mechanism of activation, which is
indicated by the conversion from a 92kDa to 82kDa isoform. However, the tissue specific
mechanisms of MMP-9 activation are not clear. In addition, the functions of different
isoforms have not been elaborated in the context of different biological processes. AIMS:
Therefore, we seek to characterize the mechanism of MMP-9 activation in human skin
during the inflammatory state, and describe the functional roles of MMP-9 in wound
healing. RESULTS: MMP-9 is expressed in human wounds and is associated with
inflammation. This observation can be partially recapitulated in multiple animal models,
showing robust MMP-9 expression and activation. Further, MMP-9 mRNA expression is
directly controlled by inflammatory cytokine, TNF-alpha, in human skin explant and cell
culture of epidermal keratinocytes. Protein expression shows similar regulation, however,
MMP-9 activation does not occur in cell culture. Incubation of purified pro-MMP-9 with
skin extract results in multiple MMP-9 cleavage products and stable MMP-9 activation.
These products can be produced in quantities that are sufficient for N-terminal Edman
degradation. CONCLUSIONS: MMP-9 activation is a prominent characteristic of
inflammation and wound healing.

1

INTRODUCTION

MMPs represent a family of several classes of matrix degrading zinc-dependent
proteolytic enzymes that are evolutionarily conserved across many organisms. Currently,
26 members of the human MMP family have been classified according to their substrates,
and together they are capable of degrading all known types of extracellular matrix (ECM)
(Page-McCaw, 2007). MMP-9 and MMP-2 are known as the gelatinases. They are
produced as latent pro-peptides and are secreted into the extracellular space where they
can bind and process several ECM components such as collagens, gelatin and
proteoglycans, and they are well known for their functions during cutaneous wound
healing (Salo, 1994) (Singer, 1999).

Inflammation is a common event in wound healing and disease and MMP-9 can augment
inflammation through the proteolytic processing and activation of latent IL-1 beta and
latent TNF-alpha (Schonbeck, 1998) (Watanabe, 1998) (Gearing, 1995). Subsequently,
these cytokines contribute to migration of keratinocytes during re-epithelialization in an
MMP-9 dependent manner (Scott, 2004) (Chen, 1995). Inflammation also activates
endothelial cells and provokes angiogenesis where MMP-9 has dual roles (Denk, 2001)
(Rajashekkar, 2006) (Macedo, 2007). MMP-9 has both pro- and anti-angiogenic
functions during wound healing (Cornelius, 1995) (Hamano, 2003) (Liu, 2008). The roles
of MMP-9 during inflammation and wound healing are pertinent to many inflammatory
diseases. Recently these connections are gathering more attention, however, cancer was

2

long ago compared to a wound that over-healed (Haddow, 1972) (Aggarwal, 2006)
(Schafer, 2008). Indeed, the regulation of MMP-9 is important to normal wound healing
and during cancer progression (Slemp, 2006) (Langers, 2008). Thus, the complex post-
translational regulation of MMP-9 is attracting increasing interest (Malla, 2008)
(Rosenblum, 2007).

The MMP-9 structure provides the backbone for post-translational regulation. MMP-9 is
made up of three domains, two terminal domains and a linker or hinge region. The N-
terminal domain is composed of the pro-peptide, three fibronectin repeats for binding
ECM proteins, and a catalytic domain. The C-terminal domain contains several
hemopexin-like repeats, which can be bound by inhibitory factors. In order to maintain its
latency, a conserved cysteine sequence (residues 97-103), PRCXXPD, in the pro-peptide
is coordinated with a zinc atom in the catalytic site. If the coordination between these
regions is disrupted, MMP-9 can expose the catalytic domain and finally liberate the pro-
peptide domain, either auto-catalytically or with the help of a protease (Rosenblum,
2007).

There are several inhibitors that can impede the activity or activation of MMP-9. Tissue
inhibitors of metalloproteinases (TIMPs) are classic regulators of MMPs, and TIMP-1
can inhibit the activity of MMP-9 by binding to the C-terminal hemopexin domain
(Wilhelm, 1989). MMP-9 can also be compartmentalized or sequestered in the
extracellular space by forming high molecular weight complexes with proteoglycans,

3

having the effect of inhibiting enzymatic activity (Malla, 2007). Alternatively, alpha-1
antichymotrypsin, an acute phase factor produced in the liver, targets the converting
enzyme and prevents the maturation of pro-MMP-9 to its active form (Han, 2008).
However, the tissue specific mechanisms of activation are not clear. Moreover, the
functional differences of the 92 and 82kDa isoforms have not been elucidated. Therefore
we propose a study wherein the specific mechanism of MMP-9 activation in human skin
can be identified, and the functional roles of "active" and latent MMP-9 in wound healing
can be elaborated.

MATERIALS AND METHODS

Materials and reagents. Cytokines were purchased from R&D Systems. Anti-vimentin
antibody and anti-type IV collagen antibody were purchased from Chemicon. Anti-pan-
cytokeratin antibody was from Dako. Anti-alpha-smooth muscle actin antibody was
purchased from Sigma. Gelatin Sepharose 4B beads were purchased from Amersham.
Bleomycin, LPS, cathepsin g, and amino phenylmercuric acetate were purchased from
Sigma. Growth media, DMEM and EpiLife were purchased from Invitrogen.

Biopsies and animals. Clinical biopsies included normal discarded skin and scars. These
were collected according to the protocol approved by the Internal Review Board at the
University of Southern California. FVB wild type mice were from Jackson Laboratory
(Bar Harbor, ME). Animal care and use was in accordance with “the Guide for Care and

4

Use of Laboratory Animals” from the National Institutes of Health and approved by the
Institutional Animal Care and Research Advisory Committee at the University of
Southern California. Mice were given two 6mm wounds by punch biopsy and either
saline (0.9%NaCl), bleomycin (Sigma) (0.5µg/µl), or LPS (Sigma) (0.5µg/µl) was loaded
directly into the wound (25µl) (n=3 mice per experimental condition). Thereafter, mice
were treated every other day for up to three weeks. When the wounds began to close,
treatments were applied by subcutaneous injection. Mice were sacrificed at 6-8 hours, 24
hours, 10 days, and 3 weeks and skin was processed for histology and real-time RT-PCR.

Cell and organ culture. Epidermal keratinocytes and dermal fibroblasts were isolated
from human skin as previously described (Garner, 1993) (Toda, 1987). Keratinocytes
were grown in complete medium (Epilife, Cascade Biologics) with streptomycin sulfate
(100 µg/ml) and penicillin (100 units/ml). Dermal fibroblasts were grown in DMEM
containing 10% fetal bovine serum with similar antibiotic. For organ culture, otherwise
discarded human skin was obtained from patients undergoing aesthetic or reconstructive
surgery in accordance to the protocol approved by the Internal Review Board at the
University of Southern California. Subcutaneous fats and connective tissue was removed
with a scalpel. Thin skin was cut into 3-5 mm
2
sections and were allowed to partially dry
until tacky. Subsequently, skin was adhered to a collagen-coated plate and growth media
was carefully added. Tissue was cultered for 5 days before adding cytokines.

5

Histology staining. Formalin-fixed skin specemins were dehydrated in alcohols, cleared
in xylene, and embedded in paraffin. Five µm-thick sections were cut and stained with
either Hematoxylin-eosin or Sirius red. Sirius red sections were later visualized under
polarized light for bifringence.

Immunofluorescence staining. Specimens were lightly fixed in formalin and embedded
in OTC compound using sucrose as a cryoprotectant. Frozen sections were cut to 7µm,
blocked by 5% non-fat milk, and incubated with anti-vimentin (MAB3400, Chemicon) at
1:200, anti-pan cytokeratin (Dako) at 1:400, anti-alpha smooth muscle actin (F3777,
Sigma) at 1:200, or anti-type IV collagen (MAB3326, Chemicon) 1:200 in 1% non-fat
milk overnight at 4
o
C. Sections were washed profusely in PBS. To detect primary
antibodies sections were incubated with FITC-conjugated-ant-mouse-IgG (for vimentin
and type IV collagen), Cy3-conjugated-anti-rabbit-IgG (for pan cytokeratin), and alpha
SMA was directly conjugated to FITC. DAPI (1µg/ml) was used to stain nuclei.

Real-time RT-PCR. Total RNA was extracted using Trizol reagent according to the
manufacturer’s instructions (Invitrogen). First strand cDNA was produced using First-
Strand cDNA Synthesis by Superscript II Reverse Transcriptase with random primers.
Two micrograms of total RNA was used for each reverse transcription reaction mixture
(20µl). Real-time PCR was carried out using the ABI Prism 7900HT (Applied
Biosystems). 10ml reactions were set in 384-well PCR plates using the final
concentrations: 1 µmol forward and reverse primers, 1x SYBR Green master mix (qPCR

6

Mastermix Plus for SYBR Green I, Eurogentec), and 5-10 ng of cDNA. For each
condition duplicates or triplicates were used to minimize variations due to pipetting error.
Cycling conditions: initial step (50
o
C for 2 min), hot activation (95
o
C for 10 min),
amplification (95
o
C for 15 s, 60
o
C for 1min) repeated 40 times, and quantification by
SYBR Green fluorescence measurement. Data were analyzed using the ABI Prism SDS
2.3 software. Gene expression was calculated by the ΔCt method, each mRNA was
normalized to GAPDH mRNA. Primers can be released upon request.

Preparation of pro-MMP-9. Transformed human keratinocytes (kindly provided from
Dr. David T. Woodley and Dr. Lily Lee at USC) were grown in EpiLife growth medium
(Cascade Biologics) to confluence. MMP-9 is secreted in the medium as the 92-kDa pro-
MMP-9 isoform. The condition media was collected and cleared by centrifugation at
4000 x g. Pro-MMP-9 was purified from conditioned media over gelatin-conjugated
Sepharose 4B column followed by washing with 400mM NaCl, 0.5% Triton X-100 in
50mM Tris, pH 7.5. The bound gelatinase was eluted using 6M urea followed by dialysis
against a buffer containing 100mM NaCl and 50mM Tris at pH 7.5 (NT buffer).

Extraction of pro-MMP-9 activator from human skin. Skin tissue was minced and
washed in NT buffer containing 100mM NaCl and 50mM Tris at Ph 7.5. Then the tissue
was extracted by 2.5% Triton X-100, 6M urea, and 2M NaCl at 4
o
C, respectively. The
extracts were subject to dialysis against NT buffer and the insoluble fractions were

7

pelleted and washed with NT. The white precipitate that formed after dialysis of the 2M
NaCl tissue extract was collected by centrifugation and re-suspended in 100µl NT buffer.

Pro-MMP-9 activation assay. For each assay 20µl of skin extracts were incubated with
10ul of pro-MMP9 together with 100mM CaCl
2
and 200µg BSA. The reaction was
carried at 37
o
C for 16 h followed by gelatinolytic zymography.

Gelatinolytic Zymography. The conditioned media and assays were loaded directly with
sample buffer into a 10% polyacrylimide gel containing 0.1% (w/v) gelatin. The gel was
run under non-reducing conditions. Electrophoresis was carried out at 60 V overnight at
room temperature. After electrophoresis the SDS in the gel was removed by incubation
with 2.5% Triton X-100. Gelatinolytic activity was developed in a buffer containing
5mM CaCl
2
, 150mM NaCl, and 50mM Tris at pH 7.5 for 16 h at 37
o
C. The gelatinolytic
activities were visualized by staining with 0.5% Coomassie Blue R-250, 10%MeOH, and
15% acetic acid.

RESULTS

Skin architecture is highly variable. Our lab is interested in the changes of tissue
architecture that have functional effects because MMPs are commonly involved.
Therefore we wished to characterize several examples in human and murine skin. Fig. 1
highlights some examples of skin diversity. 1A, H&E staining of a sample taken from a

8

single patient comparing normal skin to an adjacent hypertrophic scar. The left side of the
image represents the proximal portion of the scar. Within the scar body is an
inflammatory nodule that is directly deep to a highly convoluted rete epidermis (dark
arrow). This architecture can be compared to a normal region with a less dense dermis,
fewer inflammatory cells, and a less convoluted epidermis. The margin of the scar is
designated by the * (Bar = 300µm). 1B, immunofluorescent stain of a human
hypertrophic scar (upper) and normal skin (lower). Red (Cy3) stains a panel of
cytokeratins in the epidermis but not those in the basal epidermis. Green (FITC) reacts
with vimentin, a mesenchymal marker. The density if mesenchymal cells is higher in the
hypertrophic scar, particularly in regions near the convoluted papillary dermis (light
arrow). 1C (upper), whole mount from mice (FVB), 7-days post wounding. The wound
has become highly vascular. 1C (lower), immunofluorescent staining using the same
antibody as 1B. Light arrows indicate human dermal fibroblasts that reach into the layer
of basal keratinocytes. 1D, Hematoxylin and Eosin staining of samples taken from a
single patient comparing normal skin to an adjacent burned area that has developed into a
fiberous scar. Epidemal thickeness is much greater in burned skin and contains a thick
cornified epithelium (dark arrow) (Bar = 150µm). 1E, immunofluorescent staining of a
human scar with antibody against alpha-smooth muscle actin (primary-FITC-conjugated),
type IV collagen (FITC) and vimentin (Cy3). The light arrow indicates a vessel within
the papillary dermis. 1F, H&E staining of wound margin from mouse (FVB). The dark
arrow indicates a thickened epidermis atop a highly cellular dermis (light arrow).
Migration occurs in the direction of the light directional arrow. 1G, Sirius red staining of

9

the human hypertrophic scar shown in 1A. Magnification shows the difference in
collagens between the body of the scar and the adjacent dermis. Notice the increase in
density and disorganization within the dermis of the scar. We conclude that skin
architecture is highly variable, therefore the skin is likely a good organ for studying the
effects of MMPs in adults.

10

Figure 1. Skin architecture is highly variable

Figure 1. Skin architecture is highly variable.
(A) Hematoxylin and eosin staining of human
hypertrophic scar. Characteristics of the
epidermis and dermis differ along the length of
the scar. Tissue to the left of the asterisk is
considered hypertrophic. The dark arrow
indicates inflammatory cells. (B)
Immunofluorescent staining of a human
hypertrophic scar (upper panel) and normal skin
(lower panel). Cy3 (red) indicates reactivity with
anti-pan-cytokeratin, staining all but basal
epidermal keratinocytes. FITC (green) reacts with
anti-vimentin antibody, designating mesenchymal
cells. Notice the density of cells within the dermis
of the hypertrophic scar and the convoluted rete epidermis. The light arrow indicates an epidermal
inclusion of dermal fibroblasts. (C) Angiogenesis in a healing murine wound (upper). Vimentin-positive
dermal fibroblasts extend filipodia into basal layer of keratinocytes (lower). (D) H and E staining of human
burn injury. Patient-matched normal (left panel) and injured skin (right panel). Epidermal thickness and
keratinized epithelium differs greatly compared to normal skin. (E) Immunofluorescence of human skin.
The light arrow showing vasculature within the papillary dermis, staining green after reactivity with FITC-
conjugated anti-alpha-smooth muscle actin primary antibody. (F) H and E staining of a murine wound
margin after 7 days. Dark arrow indicates the migrating epithelial tongue. The light arrow indicates a
highly cellular dermis. (G) Siruis red stain of hypertrophic scar from (A). Notice the dense irregular
collagen bundles in the scar region.

11

Cutaneous injury is associated with the expression of inflammatory cytokines and
MMP-9. In order to understand the mechanisms that are involved with the changes in
skin architecture during wound healing we examined human scar specimens and explored
animal models for wound healing. A clear relationship between inflammatory cytokines
and MMP-9 expression has already been established, and Fig. 2 is a reiteration of this
point (Han, 2001). Patient matched skin from normal and thermal injury is shown in Fig.
2A (upper). 2A (lower), quantitative realtime RT-PCR from each sample shows
inflammatory cytokines, IL-1alpha and TNF-alpha and MMP-9 mRNA expression.
Scarred tissue contained greater number of all transcripts measured. Expression levels
were normalized by GAPDH mRNA. Some aspects can be recapitulated in animal
models for cutaneous injury and wound healing, Fig. 3. 3A, H&E staining of skin taken
from mice (FVB) after 3-week treatment. Bleomycin treatment produces a thickened
epidermis as previously described (Yamamoto, 1999). With the addition of a wound,
dermal thickness is increased substantially, and the epidermis fails to re-epithelialize
(dark arrow). LPS administration results in a thickened dermis and an intact epidermis.
Also notice the abundant fatty lobules adjacent to the wound (dark arrows). 3B,
Graphical representation of epidermal and dermal thicknesses in each model. 3C,
Quantitative realtime RT-PCR analysis of animal models. The LPS model appears to
evoke a robust response by the induction of TLRs, MMPs, and inflammatory cytokines.
3D, Gelatinolytic zymography. Additionally, the MMP signature from all murine wounds
exhibited many MMPs, including active MMP-9 and MMP-2. We can conclude that
cutaneous injury is associated with inflammatory cytokines and MMP-9 expression,

12

which can be recapitulated in animal models. Furthermore, zymography shows a MMP
signature from an animal wound that is similar to that reported in humans (Wysocki,
1993).

Figure 2. Cutaneous injury is associated with inflammatory cytokines and MMP-9

Figure 2. Cutaneous injury is associated with inflammatory cytokines and MMP-9. (Upper) Patient-
matched normal and burned skin. (Lower) Realtime RT-PCR analysis. mRNA was normalized by GAPDH
mRNA.

13

Figure 3. Animal models of cutaneous injury and wound healing recapitulate some
characteristics of human wound healing

14

Figure 3. Animal models of cutaneous injury and wound healing recapitulate some aspects of human
wound healing. (A) Hematoxylin and Eosin staining of murine skin after 3-week treatment as indicated by
the left and right panels. Bleomycin is a chemical toxin known to promote skin injury and fibrogenesis.
Lipopolysaccharide (LPS) is derived from gram-negative bacterial cell wall and stimulates innate immunity
through interaction with toll-like receptor 4 (TLR-4). (B) Dermal and epidermal thicknesses measured
using Adobe Photoshop©. (C) Realtime RT-PCR analysis of murine skin after wounding and treatment
indicated by the x-axis. Markers of innate immunity are increased in conjunction with MMPs. All readings
were normalized by GAPDH mRNA. (D) MMP wound signature. Murine skin was homogenized and
MMPs were captured on gelatin beads in DMEM for 1 hour. A panel of MMPs are induced. The 82kDa
band indicates active MMP-9.

15

MMP-9 expression is controlled by inflammatory cytokines and growth factors, but
MMP-9 activation does not occur in cell culture. In order to understand how MMP-9 is
regulated in human skin we treated skin explant and epidermal keratinocytes with
cytokines and growth factor. As previously described, MMP-9 expression is controlled
by inflammatory cytokines and growth factors during wound healing, Fig. 4 (Han, 2001)
(Han, 2005). 4A, Quantitative realtime RT-PCR analysis shows MMP-9 expression in
human skin explant treated with inflammatory cytokines and growth factor for 6-8 hours.
TNF-alpha is a potent activator of MMP-9 expression and this effect is augmented
synergistically by TGF-beta. MMP-9 mRNA was normalized by GAPDH mRNA.
Zymographic analysis revealed activation of MMP-9 (data not shown). 4B, Gelatin
zymography. In normal human epidermal keratinocytes MMP-9 protein responds in a
similar manner to mRNA, however, activation of pro-MMP-9 does not occur. In
conclusion, cell culture cannot reliable recapitualtio MMP-9 activation, which is
observed in tissue culture and during in vivo wound healing.

16

Figure 4. During wound healing, MMP-9 is controlled by inflammatory cytokines
and growth factors

Figure 4. During wound healing, MMP-9 is controlled by inflammatory cytokines and growth
factors. (A) Realtime RT-PCR analysis. Whole skin explant was treated with inflammatory cytokines and
growth factor for 6-8 hours. (B) Gelatin zymography from the conditioned media of normal human
epidermal keratinocytes. 18 hour treatment with inflammatory cytokine and growth factor have a similar
effect on the induction of MMP-9 protein in normal human epidermal keratinocytes. Notice the apparent
lack of MMP-9 activation in cell culture.

17

Epidermal keratinocytes and dermal fibroblasts provide different cytokines and
MMPs to the wound envionment in response to TNF-alpha. After noticing the
discrepancy in MMP-9 activation between tissue and cell culture we wished to
characterize the responses of epidermal keratinocytes and dermal fibroblasts to
inflammatory cytokine, Fig. 5. We noticed that these cells respond differently to
stimulation with TNF-alpha. All measurements were normalized by GAPDH mRNA.
Epidermal keratinocytes produce greater amounts of TNF-alpha MMP-9 and MMP-13
transcripts when compared to fibroblasts, which produce greater IL-1a mRNA in
response to TNF-alpha. As a control, we also measured E-cadherin, a cell adhesion
molecule specific to epithelial lineage. In response to TNF-alpha, E-cadherin expression
is reduced in epidermal keratinocytes. The expression of E-cadherin in fibroblasts is low,
consistent with this mesenchymal lineage. We conclude that keratinocytes are the
primary target of TNF-alpha in human skin, and response much differently than
fibroblasts, highlighting an element of complexity that is lost in monoculture.

18

Figure 5. Epidermal keratinocytes and dermal fibroblasts provide different
cytokines and MMPs to the wound environment in response to TNF-alpha

Figure 5. Epidermal keratinocytes and dermal fibroblasts provide different cytokines and MMPs to
the wound environment in response to TNF-alpha. Real-time RT-PCR analysis of epidermal
keratinocytes and dermal fibroblasts after treatment with TNF-alpha for 6 hours. Epidermal keratinocytes
contribute the majority of reciprocal TNF-alpha and MMPs while dermal fibroblasts contribute IL-1alpha.
Epidermal keratinocytes lose E-cadherin expression which might correlate to a migrative phenotype.

19

MMP-9 is activated by multiple mechanisms. In order to understand the mechanism of
MMP-9 activation we wished to use a reliable in vitro system for MMP-9 activation.
Therefore, we explored several mechanisms of activating purified pro-MMP-9 shown in
Fig. 6. Activation of pro-MMP-9 requires the liberation of the pro-domain shown in
yellow, Fig. 6A. Amino-phenylmercuric acetate can activate MMP-9 by disruption of the
coordination between a conserved cysteine and zinc ion, Fig. 6B. Several cleavage
products can be identified by gelatin zymography and the gelatinolytic activity is greatly
enhanced during the processing of pro-MMP-9. Human inflammatory neutrophils
produce a serine protease, cathepsin G, which has also been shown to activate MMP-9 in
vitro, Fig 6C (Han, 2008). Fig 6D, We prepared skin extract from normal human skin
which contains a partially purified MMP-9 activator. Insoluble material that forms after
dialysis against NT buffer at 4
o
C has the greatest crude converting activity after it is
pelleted and resuspended in NT. In conclusion, skin extract represents a reliable and
stable method of pro-MMP-9 activation.

20

Figure 6. MMP-9 is activated by multiple mechanisms

Figure 6. MMP-9 is activated by multiple mechanisms. (A) Space-filling model of the catalytic and pro-
domain of MMP-9 (Elkins, 2002). Coordination of a conserved cysteine in the pro-domain maintains the
latency of pro-MMP-9. (B) Gelatin zymography. Disruption of the coordination of the conserved cysteine
by amino-phenylmercuric acetate (APMA) results in autocleavage and subsequent activation of MMP-9.
(C) Incubation with cathepsin G, secreted by human neutrophils, results in MMP-9 activation. (D) A local
MMP-9 activator can be partially purified from human skin and used to activate MMP-9 in vitro.

21

The activating enzyme might be a high molecular weight or a conjugate protein. To
learn about the activity of the skin extract we performed a crude characterization. Fig.
7A, Gelatinolytic zymography. Activation of MMP-9 by skin extract occurs in a stepwise
manner that was recently described (Rosenblum, 2007). 7B, Of the three crude protein
extracts collected from human skin, NaCl contained the most efficient MMP-9 converting
activity. This activity could be enhanced using a 50MWCO protein concentrator while
the material that passed through the membrane contained little to no converting activity.
Furthermore, after washing with NT, the activity was retained. Fig. 7D, The MMP-9
cleavage products after overnight incubation with skin extract can be visualized after re-
purified, resolved by electrophoresis, blotted to a membrane and stained with coomassie
blue. Pro-MMP-9, the intermediate cleavage product, and active MMP-9 can be
identified by comparing the relative ratio between bands stained by coomassie blue to
zymography and also aligning the apparent molecular weight, 7E (light and dark arrows).
Skin extract alone does not have gelatinolytic activity measured by zymography (data not
shown). We conclude that the activating enzyme (if >50kDa) or complex is extracted on
a charge-based attraction. Also, this in vitro method may be sufficient to study the
cleavage products of pro-MMP-9, but we should determine which bands are derived from
skin extract alone.

22

Figure 7. Skin extract contains the activating enzyme and can be used to study the
mechanism of MMP-9 activation in vitro

Figure 7. Skin extract
contains the activating
enzyme and can be used to
study the mechanism of
MMP-9 activation in vitro.
(A) Gelatin zymography.
Activation of MMP-9 by skin
extract occurs in a stepwise
fashion. (B) Converting activity
in skin extract is isolated to the
NaCl fraction, and can be
retained by a 50 MWCO
protein concentrator (upper
panel). After washing the 50
MWCO retained material three
times, the activity is preserved (lower panel). (C) Coomassie blue staining of purified pro-MMP-9 after
transfer to a PVDF membrane. (D) Cleavage products after incubate with skin extract and re-purification
and concentration using the same methods as in Fig. 8C. (E) By comparison of zymography and coomassie
blue staining we can identify presumptive MMP-9 isoforms. These fragments can be characterized by N-
terminal Edman degradation and used to characterize and finally purify the activating enzyme.

23

The MMP-9 cleavage products from skin extract can be prepared for protein
sequencing. The purpose of this study is to characterize the MMP-9 activating enzyme in
human skin. Ultimately, we wish to identify the site of MMP-9 cleavage that occurs
during processing, as to will allow us to purify the converting enzyme using a direct
approach by designing a small peptide that will act as a pseudo- suicide substrate.
Therefore, we frequently correspond with the proteomics department at the University of
Texas. Currently, we have identified problems with the method of protein purification
and we will adjust our protocol slightly before resubmission of protein for sequencing.

DISCUSSION
Human skin represents a model organ for the study of MMPs. Figure 1G demonstrates
that matrix biology contributes to skin diversity and the outcomes of inflammation and
wound healing. There are several clinical examples of dysfunctions in inflammation and
wound healing that lead to skin fibrosis or hypertrophic and keloid scars. These
dysfunctions of matrix regulation are the motivating drive of our lab. Previously, we have
characterized multiple mechanisms of MMP-9 transcriptional and post-translational
regulation in human skin (Han, 2001) (Han, 2002) (Han, 2005) (Han, 2008). Now, our
aim is to identify the pro-MMP-9 converting enzyme in human skin.

We chose this aim for multiple reasons. One reason is that human skin continues to evade
our understanding. Presently, we are still attempting to develop adequate models for
wound healing. So far, animal models have fallen somewhat short, and some require

24

elaborate methods or special facilities. Our brief exploration of the bleomycin and LPS
models support this notion. Originally, we explored these animal models as having the
potential to recapitulate hypertrophic scarring, characterized by over abundantly
produced ECM. The bleomyin model was previously described to produce skin fibrosis
so it seemed logical that we could extend this model to wound healing and hypertrophic
scarring (Yamamoto, 1999). We found that it an inappropriate model for hypertrophic
scarring, however, because it left the epithelium severely damaged Fig. 3. Due to the
challenges to reproduce elements of human wound healing in animals, there is still much
to be learned of dysregulated human wound healing. MMPs represent a good candidate
for studying regulation of the wound environment because they have been described to
modulate inflammation and also predict the survivability of some cancers (Manicone,
2008) (Langers, 2008). Therefore, after identifying of the converting enzyme we could
gain a better understating of how to perturb the wound environment. The second reason
we seek to identify the pro-MMP-9 converting enzyme is because it represents an aspect
of skin biology that cell culture cannot reproduce. Understandably, reduction of a tissue
to its cellular constituents has favored many scientific discoveries. However, the absence
of MMP-9 activation illustrates a void in cell culture that reliably occurs in tissue.
Identification of the MMP-9 activator will contribute to the understanding of wound
healing, but it might also help us develop better environments for cell culture, which may
contain more defined ECM. After identification of the activating enzyme, we might be
able to reliably recapitulate the characteristic MMP-9 activation in cell culture.

25

Originally, the goal of this study was to identify the cleavage site of pro-MMP-9. We
considered using commercial sources as pro-MMP-9 activators, but with the ultimate
goal of identifying the converting enzyme these methods may only lead us back to the
commercial source. Therefore, we must us skin extract to convert pro-MMP-9. In
addition, it appears that the mechanism of activation could be more complicated than
anticipated. Recall, the activating enzyme was extracted from skin on a charge-based
attraction. Additionally, the activity of skin extract was retained in a 50kDa MWCO
protein centrifugal concentrator even after being washed numerous times in NT buffer.
Recently, we performed an additional experiment and washed skin extract consecutively
with 0.1% and 0.5% SDS. If the converting enzyme is actually less than 50kDa but exists
in a complex that allows it to be retained in the concentrator, then washing with SDS
might allow the converting enzyme to pass through. After washing with 0.1% SDS,
incubation of the retained skin extract with purified pro-MMP-9 abolished MMP-9
activity altogether, it could not be detected by zymography. The passed material did not
appear to have converting activity. Interestingly, after washing with 0.5% SDS,
incubation of the retained skin extract with pro-MMP-9 restored the gelatinolytic activity.
Furthermore, the retained material had pro-MMP-9 converting activity (data not shown).
These results suggest that the converting enzyme could represent a complex of proteins
that are extracted based on charge, but can be disrupted and thereby modulate their
function. By continuing to study this mechanism of activation, we hope to elucidate the
MMP-9 activator, and the functional roles in human skin.

26

References

Chen, J. D., Lapiere, J. C., Sauder, D. N., Peavey, C., Woodley, D. T. (1995) Interleukin-
1 alpha stimulates keratinocyte migration through an epidermal growth
factor/transforming growth factor-alpha-independent pathway. J. Invest. Dermatol. 104,
729-733

Cornelius, L. A., Nehring, L. C., Roby, J. D., Parks, W. C., Welgus, H. G. (1995) Human
dermal microvascular endothelial cells produce matrix metalloproteinases in response to
angiogenic factors and migation. J. Invest. Dermatol. 105, 170-176

Denk, A., Goebeler, M., Schmid, S., Berberich, I., Ritz, O., Lindemann, D., Ludwig, S.,
Wirth, T. (2001) Activation of NF-kappaB via IkappaB kinase complex is both essential
and sufficient for proinflammatory gene expression in primary endothelial cells. J. Biol.
Chem. 276, 28451-28458

Elkins, P. A., Ho, Y. S., Smith , W. W., Janson, C. A., D’Alessio, K. J., McQueney, M.
S., Cummings, M. D., Romanic, A. M. (2002) Structure of the C-terminally truncated
human ProMMP-9, a gelatin-binding matrix metalloproteinase. Acta. Crystallogr. D.
Biol. Crystallogr. 58, 1182-1192

Garner, W. L., Karmiol, S., Rodriguez, J. L., Smith, D. J., Phan, S. H. (1993) Phenotypic
differences in cytokine responsiveness of hypertrophic scar verses normal dermal
fibroblasts. J. Invest. Dermatol. 101, 875-879

Gearing, A. J., Beckett, P., Christodoulou, M.,. Churchill, M., Clements, J. M., Crimmin,
M., Davidson A. H., Drummond, A. H., Galloway, W. A., Gilbert, R. (1995) Matrix
metalloproteinases and processing of pro-TNF-alpha. J. Leuk. Biol. 57, 774-777

Haddow, A. (1972) Molecular repair, wound healing, and carcinogenesis: tumor
production a possible overhealing? Adv. Cancer Res. 16, 181-234

Hamano, Y., Zeisberg, M., Sugimoto, H., Lively, J. C., Maeshima, Y., Yang, C., Hynes,
R. O., Werb, Z., Sudhakar, A., Kalluri, R. (2003) Physiological levels of tumstatin, a
fragment of collagen IV alpha3 chain, are generated by MMP-9 protolysis and suppress
angiogenesis via alphaV beta3 integrin. Cancer Cell 3, 513-514

Han, Y. P., Yan, C., Garner, W. L. (2008) Proteolytic activation of matrix
metalloproteinase-9 in skin wound healing is inhibited by alpha-1-antichymotrypsin. J.
Invest. Dermatol. Apr 10 (Epub ahead or print)

27

Han, Y. P., Downey, S., Garner, W. L. (2005) Interleukin-1-alpha-induced proteolytic
activation of metalloproteinase-9 by human skin. Surgery 138, 932-939

Han, Y. P., Tuan, T. L., Wu, H., Hughes, M., Garner, W. L. (2001) Transforming growth
factor-beta and tumor necrosis factor-alpha-mediate induction and proteolytic activation
of MMP-9 in human skin. J. Biol. Chem. 276, 22341-22350

Inkinene, K., Turakainen, H., Wolff, H., Ravanti, L., Kahari, V. M., Ahonen, J. (2000)
Expression and activity of matrix metalloproteinase-2 and -9 in experimental granulation
tissue. APMIS 108, 318-328

Langers, A. M., Sier, C. F., Hawinkels, L. J., Kubben, F. J., van Duijn, W., van der
Reijden, J. J., Lamers, C. B., Hommes, D. W., Verspaget, H. W. (2008) MMP-2 geno-
phenotype is prognostic for colorectal cancer survival, whereas MMP-9 is not. Br. J.
Cancer 98, 1820-1823

Liu, H., Chen, B., Lilly, B. (2008) Fibroblasts potentiate blood vessel formation partially
through secreted factor TIMP-1. Angiogenesis Feb 14 (Epub ahead of print)

Macedo, L., Pinhal-Enfield, G., Alshits, V., Elson, G., Cronstein, B. N., Leibovich, S. J.
(2007) Wound healing in impaired in MyD88-deficient mice: a role for MyD88 in the
regulation of wound healing by adenosine A2A receptors. Am. J. Pathol. 171, 1774-1788
Scott, K. A., Arnott, C. H., Robinson, S. C., Moore, R. J., Thompson, R. G., Marshallm J.
F., Balkwill, F. R. (2004) Tnf-alpha regulates epithelial expression of MMP-9 and
integrin alphavbeta6 during tumor promotion. A role for TNF-alpha in keratinocyte
migration? Oncogene 23, 6054-6966

Malla, N., Berg, E., Lars, U. H., Jan-Olof, W. (2007) Interaction of pro-matrix
metalloproteinase-9/proteoglycan heteromer with gelatin and collagen. J. Biol. Chem.
283, 13652-13665

Manicone, A. M., McGuire, J. K. (2008) Matrix metalloproteinses as modulators of
inflammation. Sem. Cell Dev. Biol. 19, 34-41

Page-McCaw, A., Ewald, A. J., Werb, Z. (2007) Matrix metalloproteinases and the
regulation of tissue remodeling. Nat. Rev. Mol. Cell Biol. 8, 221-233

Rajashekkar, G., Willuweit, A., Patterson, C. E., Sun, P., Hilbig, A., Breier, G., Helisch,
A., Clauss, M. (2006) Continuous endothelial cell activation increases angiogenesis:
evidence for the direct role of endothelium linking angiogenesis and inflammation. J.
Vasc. Res. 43, 193-204

28

Rosenblum, G., Van den Steen P. E., Cohan, S. R., Grossman, J. G., Frenkel, J.,
Sertchook, R., Slack, N., Strange, R. W., Openakker, G., Sagi, I. (2007) Insights intot eh
structure and domain flexibility of full-length pro-matrix metalloproteinase-9/gelatinase
B. Structure 15, 1159-1161

Rosenblum, G. Meroueh, S. Toth, M., Fischer, J. F., Fridman, R., Mobashery, S., Sagi, I.
(2007) Molecular structures and dynamic of the stepwise activation mechanism of a
matrix metallorproteinase zymogen: challenging the cysteine switch dogma. J. Am.
Chem. Soc. 129, 13566-13574

Schafer, M., Werner, S. (2008) Cancer as an overhealing wound: an old hypothesis
revisited. Nat. Rev. Mol. Cell Biol. 9, 628-638

Schonbeck, U., Mach, F., Libby, P. (1998) Generation of biologically active IL-1 beta by
metalloproteinases: a novel caspase-1-independent pathway of IL-1 beta processing. J.
Immunol. 161, 3340-3346

Shepard, P., Martin, G., Smola-Hess, S., Brunner, G., Kried, T., Smola, H. (2004)
Myofibroblast differentiation is induced in keratinocyte-fibroblast co-cultures and is
antagonistically regulated by endogenous transforming growth factor-beta and
interleukin-1. Am. J. Pathol. 164, 2055-2066

Slemp, A. E., Kirschner, R. E. (2006) Keloids and scars: a revier of keloids and scars,
their pathogenesis, risk factors, and management. Curr. Opin. Pediatr. 18, 396-402

Toda, K., Tuan, T. L., Brown, P. J., Grinnell, F. (1987) Fibronectin receptors of human
keratinocytes and their expression during cell culture. J. Cell Biol. 105, 3097-3104

Watanabe, N., Nakada, K., Kobayashi, Y. (1998) Processing and release of tumor
necrosis factor alpha. Eur. J. Biochem. 253, 576-582

Wysocki, A. B., Staiano-Coico, L., Grinnell, F. (1993) Wound fluid from chronic leg
ulcers contains elevated levels of metalloproteinases MMP-2 and MMP-9. J. Invest.
Dermatol. 101, 64-68

Yamamoto, T., Takagaw, S., Katayama, I., Yamazaki, K., Shinkai, H., Nishioka, K.
(1999) Animal models of sclerotic skin. I: Local injections of bleomycin induce skin
mimicking scleroderma. J. Invest. Dermatol. 112, 456-462

Abstract (if available)

Abstract BACKGROUND: Matrix metalloproteinases (MMPs) represent a family of zinc-dependent proteases that contain a conserved pro-peptide domain. MMP-9 is a gelatinase known for degrading type IV collagen among its potential substrates. MMP-9 is implicated in a myriad of different biological processes and pathologies such as wound healing, and cancer. Thus, strict transcriptional and post-translational regulation of MMP-9 is important. Latent pro-MMPs have a conserved mechanism of activation, which is indicated by the conversion from a 92kDa to 82kDa isoform. However, the tissue specific mechanisms of MMP-9 activation are not clear. In addition, the functions of different isoforms have not been elaborated in the context of different biological processes. AIMS: Therefore, we seek to characterize the mechanism of MMP-9 activation in human skin during the inflammatory state, and describe the functional roles of MMP-9 in wound healing. RESULTS: MMP-9 is expressed in human wounds and is associated with inflammation. This observation can be partially recapitulated in multiple animal models, showing robust MMP-9 expression and activation. Further, MMP-9 mRNA expression is directly controlled by inflammatory cytokine, TNF-alpha, in human skin explant and cell culture of epidermal keratinocytes. Protein expression shows similar regulation, however, MMP-9 activation does not occur in cell culture. Incubation of purified pro-MMP-9 with skin extract results in multiple MMP-9 cleavage products and stable MMP-9 activation. These products can be produced in quantities that are sufficient for N-terminal Edman degradation. CONCLUSIONS: MMP-9 activation is a prominent characteristic of inflammation and wound healing.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Histone deacetylase 4 represses matrix metalloproteinases in myofibroblastic hepatic stellate cells

PDF

Mechanisms of human skin cell migration and wound healing

PDF

The roles of type VII collagen in wound healing and scar reduction

PDF

Molecular mechanism of transforming growth factor-beta signaling in skin wound healing

PDF

Determining the necessity of nuclear matrix metalloproteinase-2 (MMP-2) activity for proficient skeletal muscle differentiation

PDF

Characterizing the role of Dnmt1 and its downstream factors in progressive alopecia

PDF

A novel role for hypoxia-inducible factor-1alpha (HIF-1alpha) in the regulation of inflammatory chemokines and leukotriene expression in sickle cell disease

PDF

Exploring alternative roles of visual arrestin 1 in photoreceptor synaptic regulation and deciphering the molecular pathway of retinal degeneration using mouse knockout technology

PDF

Renal extracellular matrix bioactivity in a range of physiological states: novel insights for kidney regeneration

Asset Metadata

Creator Grimm, Wesley Andrew (author)

Core Title Regulation of matrix metalloproteinase-9 in cutaneous wound healing

School Keck School of Medicine

Degree Master of Science

Degree Program Experimental and Molecular Pathology

Publication Date 11/20/2008

Defense Date 08/13/2008

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag matrix metalloproteinase,MMP,MMP activation,MMP9,MMP-9,OAI-PMH Harvest,wound healing

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Han, Yuan-Ping (committee chair), Chuong, Cheng-Ming (committee member), Garner, Warren (committee member)

Creator Email wes.grimm@gmail.com,wgrimm@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1809

Unique identifier UC1101138

Identifier etd-Grimm-2402 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-128121 (legacy record id),usctheses-m1809 (legacy record id)

Legacy Identifier etd-Grimm-2402.pdf

Dmrecord 128121

Document Type Thesis

Rights Grimm, Wesley Andrew

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu