Published online before print March 11, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,1
* Center for Lung Biology and Institute for Stem Cell and Regenerative Medicine and Departments of
Medicine (Pulmonary and Critical Care Medicine),
Pathology, and
Pediatrics (Critical Care Medicine), University of Washington, Seattle, Washington, USA
1Correspondence: Center for Lung Biology, University of Washington, 815 Mercer Street, Seattle, WA 98109, USA. E-mail: parksw{at}u.washington.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: inflammation • epithelium • protease
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Neutrophil influx into wounded or infected tissue is stimulated and guided, to a large part, by an activated epithelium, and despite their specialization to serve distinct functions in different tissues, epithelia respond similarly to injury and regulate local inflammation by equivalent mechanisms. Following injury, epithelial cells initiate a programmed series of coordinated responses, such as proliferation and migration, to restore tissue integrity. By their production of chemoattractants, adhesion molecules, and other proteins, epithelial cells recruit and confine the influx of inflammatory cells to sites of injury. Although seemingly divergent events, the epithelial programs regulating repair and inflammation, as well as other innate defense mechanisms, may have coevolved, particularly the proteins that serve essential functions in these processes. Hence, many of the epithelial products associated with any one of these events are likely common to all. Although long thought to function primarily in the turnover and degradation of the extracellular matrix, matrix metalloproteinases (MMPs) have emerged as a family of extracellular processing enzymes that control the activity of several diverse proteins with predominant roles in repair and inflammation [18 ].
Studies by us and others [19
20
21
] indicate that matrilysin (MMP7), which is expressed almost exclusively by mucosal epithelia, functions in various processes of innate immunity. For example, matrilysin is highly expressed in infected tissue [22
], and its expression is markedly up-regulated by exposure to gram-negative bacteria [23
, 24
]. In mice, matrilysin activates intestinal pro-
-defensins (cryptdins), a family of antimicrobial peptides, and as a result of the lack of mature -defensins, Mmp7–/– mice have an impaired ability to battle enteric pathogens [25
]. Matrilysin also sheds membrane-bound Fas ligand [26
], and its ability to promote apoptosis may be another point of action for this proteinase in innate immunity [18
]. Matrilysin is also prominently expressed by migrating epithelium in injured airway [27
], suggesting an additional function in epithelial repair. Indeed, whereas wounds made in wild-type trachea repair rapidly, Mmp7–/– tissue shows no evidence of epithelial migration, even several days post-injury [27
, 28
]. Furthermore, by its ability to shed cell surface proteoglycans, matrilysin affects chemokine and growth factor activity [29
, 30
].
To test further the role of matrilysin in response to tissue injury, we studied the response of Mmp7–/– mice subjected to acute colon injury. We focused on matrilysin in this model, as it is prominently expressed by epithelial cells adjacent to ulcers in patients with inflammatory bowel disease [31 ] and by the inflamed mucosa of IL-10 null mice [32 ]. Our findings in mice indicate that matrilysin mediates beneficial and deleterious effects in response to tissue injury. On one hand, matrilysin is required for repair of damaged epithelium; on the other, matrilysin controls the transepithelial influx of neutrophils via its ability to affect chemokine gradients, which if excessive, can lead to severe tissue damage. We propose that matrilysin functions as a checkpoint barring inappropriate neutrophil activation at the mucosal surface.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Quantitative PCR
Total RNA was isolated with Trizol (Invitrogen, Carlsbad, CA, USA). Primers and TaqMan probes (FAM dye-labeled) for MMP7, 1(I) collagen, and GAPDH were added to cDNA and synthesized from 5 µg total RNA with a High-Capacity cDNA Archive kit (Applied Biosystems, Foster City, CA, USA), and product amplification was measured with an ABI HT7900 Fast Real-Time PCR system. The threshold cycle (Ct) was obtained from duplicate samples and averaged, and the Ct range for each is shown in the figure legends. The 
Ct was the difference between the average Ct for the specific cDNAs and the average Ct for GAPDH, which ranged from 16.97 to 14.16. The Ct was the average Ct at a given time-point minus the average Ct of Day 0 samples. The data are expressed as relative quantification (RQ), which is the fold change, and calculated as 2–Ct.
Histology
Washed colons were fixed in 10% formalin overnight and then dehydrated in 70% ethanol. Colons were opened longitudinally and rolled with the mucosa facing outwards using a modified Swiss-roll technique [34
]. Intestinal rolls were dipped into a cryomold containing 5% formalin and 2% agarose and processed for routine paraffin embedding. Sections (4 µm) were stained with H&E or processed for immunohistochemistry or in situ hybridization. To quantify injury, we determined the crypt scores [33
]. This morphometric analysis presents injury as the percent of the total colon length with a specific grade of lesion, from 0 (unaffected) to 4 (most severe). Specifically, the grading was as follows: Grade 0, intact crypts; Grade 1, loss of basal one-third of the crypt; Grade 2, loss of basal two-thirds of the crypt; Grade 3, loss of entire crypt but surface epithelium intact; and Grade 4, loss of entire crypt including surface epithelium. Samples were assessed in a blinded manner by two authors. Images were captured using an Olympus BX-51 fluorescence/differential interference contrast (DIC) microscope with U plan Apo 1.25x/0.04, 2x/0.05, 10x/0.3, and 20x/0.70 objectives and an Olympus DP25 5.5 megapixel digital camera.
Immunohistochemistry/in situ hybridization
Serial sections were deparaffinized, immersed in 10 mM citrate buffer (pH 6.0) at 98°C for 40 min for antigen retrieval, and processed for immunostaining using Vectastain ABC Elite kits (Vector Laboratories, Burlingame, CA, USA) as described [27
]. Matrilysin was detected with a rat anti-human mAb [35
]. Neutrophils were identified with MCA771GA (1:1,000, Serotec, Oxford, UK), and macrophages were stained for F4/80 (1:50; MCAP497, Serotec). Controls were processed with matching preimmune sera or nonimmune ascites. In vitro-transcribed antisense and sense RNA probes for matrilysin were labeled with [35S]-UTP, and sections were hybridized as described [36
].
Protein assays
Myeloperoxidase (MPO) levels were determined by ELISA (HyCult Biotech, The Netherlands). Colon tissue was homogenized in 2.5 g hexadecyltri-methylammonium bromide buffer in 500 ml Milli-Q water. The homogenate was centrifuged at 8000 g, and the supernatant was collected. For chemokines and cytokines, colon samples were homogenized in 2 ml 150 mM NaCl, 15 mM Tris (pH 7.2), 1 mM MgCl2, 1 mM CaCl2, 1% Triton X-100, and Complete Miniproteinase inhibitor cocktail (Roche Molecular Biochemicals, Indianapolis, IN, USA). Homogenates were centrifuged at 10,000 g for 15 min at 4ºC, and the supernatants were collected. Reagents used for the Luminex assay were purchased from R&D Systems (Minneapolis, MN, USA) and include the mouse Fluorokine MAP kit for keratinocyte-derived chemokine (KC), MIP-2, MCP-1, TNF-, IL-6, IL-12, IFN-
, IL-1β, IL-4, and IL-10. KC and MIP-2 levels in colon lavages were measured using Quantikine ELISA kits from R&D Systems.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (14K): [in a new window] |
Figure 1. Sustained injury in Mmp7–/– mice. (A) Weight (Wt) change was plotted as percent of weight at Day 0 for wild-type (n=37) and Mmp7–/– (n=32) mice. The arrow indicates the cessation of DSS at Day 7. Data are the mean ± SEM of three separate experiments. (B, C) Colon length and weight were measured after the start of DSS exposure. Data are the mean ± SEM of six samples/genotype/day. Length: P <0.05 at Days 15 and 21; weight: P = 0.0033 at Day 21. (D) Survival of wild-type (n=132; 53 females and 79 males) and Mmp7–/– (n=110; 43 females and 67 males) mice was followed for 21 days in eight experiments; P < 0.0001.
|
1(I) collagen mRNA, used as a marker of interstitial scarring, and total tissue protein did not differ significantly between genotypes at any time-point (data not shown). Although we saw only a small, yet significant, difference in weight loss between genotypes during the onset of colitis (Days 0–10), susceptibility to the lethal effects of DSS-induced injury was much more pronounced in wild-type than in Mmp7–/– mice (Fig. 1D) . Whereas 73% of wild-type mice (n=132 at Day 0) died between 8 and 12 days after the onset of 2.5% DSS, which agrees with the lethality noted for C57BL/6 mice by others [39 , 40 ], only 32% of Mmp7–/– mice (n=110) succumbed over this period (Fig. 1D) . No additional mice of either genotype died after Day 15. Resistance to lethality in Mmp7–/– mice was not a result of reduced consumption of DSS–water. Over the 7 days of treatment, Mmp7–/– mice drank slightly more DSS–water (2.03±0.26 l/g body weight) than did wild-type mice (1.76±0.23 l/g). Although this small difference reached statistical significance (P=0.044; n=12), it did not correlate with mortality.
Persistent injury in Mmp7–/– mice
To dissect further the role of matrilysin, we examined sections of colons for degree of injury and inflammation (Fig. 2
). In both genotypes, tissue injury was most severe in the middle colon, less in the distal colon, and least in the proximal colon. Using an established injury-scoring method [33
], based largely on epithelial crypt morphology, less-severe Grades 1 and 2 lesions dominated the early stages (Days 4–7) after onset of DSS and were more prevalent and appeared earlier in wild-type mice than in Mmp7–/– mice. At Day 4, 24.7% ± 5 of the mucosal surface of wild-type colons had type 1 focal injuries and 5.5% ± 1 as type 2 lesions. In contrast, only 7% ± 2 of Mmp7–/– colons showed damage, and essentially, all of these focal injuries were classified as type 1 lesions. At Day 7, more Grade 1 injuries were still evident in wild-type colons, but the extent of Grades 2, 3, and 4 injuries did not differ between genotypes.
 View larger version (72K): [in a new window] |
Figure 2. Delayed yet persistent ulceration in Mmp7–/– colons, which from wild-type (WT) and Mmp7–/– mice were harvested at Days (D) 4, 7, 11, and 21 post-DSS and examined by H&E staining. Several areas are numbered indicating the crypt score assigned to that site. Original bar = 2 mm.
|
Fig. 6
). By Day 21, normal crypt architecture was mostly restored in the wild-type mice, but large patches of epithelial erosion, crypt loss, and abnormal crypt structures (Grades 2–4 lesions) remained in the knockout mice.
 View larger version (23K): [in a new window] |
Figure 3. Matrilysin expression. (A) At the indicated days post-onset of 2.5% DSS, colons from wild-type mice were harvested and divided into proximal (P) and distal (D) halves. Total RNA was isolated, and matrilysin and GAPDH mRNA levels were assessed by Northern hybridization. Autoradiograms from three experiments were scanned (representative blots are shown), and the densitometric values for matrilysin mRNA were normalized to those for GAPDH mRNA and averaged. (B) Total RNA was isolated from colons of wild-type mice (n=6) at various days after the onset of exposure to 2.5% DSS, and mRNA levels were detected by quantitative RT-PCR. The data are expressed as RQ, as described in Materials and Methods. The Ct range for the MMP7 probe set among experiments was 27.91–23.95.
|
 View larger version (173K): [in a new window] |
Figure 4. Matrilysin expression in DSS-induced colon injury. By immunohistochemistry, matrilysin protein was detected in the mucosal folds of the proximal colon (A), within regenerating epithelial (B–E). Using in situ hybridization of serial sections, matrilysin mRNA was also detected in regenerating areas and near ulcerations (D', arrowheads), but the location of the mRNA-positive cells did not always overlap with the protein-positive cells (*). However, several examples of a coincident signal for matrilysin protein and mRNA were seen, often in the mid-to-basal crypt areas (D, D', E, and E', thick arrows). Original bars = 100 µm.
|
 View larger version (125K): [in a new window] |
Figure 5. Delayed and restrained neutrophil influx at Days 4 and 7 in Mmp7–/– mice. Colons from wild-type and Mmp7–/– mice at Days 4 and 7 were harvested and embedded, and neutrophils were identified by staining with antibody MCA771GA (brown). The boxed areas in the large micrographs (D4 and D7) were captured under higher magnification with DIC optics and are displayed below. Arrows mark the mucosal–submucosal interface. Original bars = 2 mm (D4, D7); 100 µm (D7'); 50 µm (insets).
|
 View larger version (126K): [in a new window] |
Figure 6. Impaired re-epithelialization in Mmp7–/– colons. Sections of colons from wild-type and Mmp7–/– (knockout) mice at Days 15 and 18 were stained for neutrophils. Ongoing (D15) and completed (D18) re-epithelialization (arrows) was seen over ulcers in wild-type colons, restrained epithelial migration at the edge of wounds, and persistent ulcerations were common in Mmp7–/– colons. Original bars = 100 µm.
|
The presence of staining in epithelial cells at the base of and along the length of the crypts and on the mucosal surface between crypts (Fig. 4A 4B 4C 4D) suggested that matrilysin was produced by basal epithelial cells that had migrated up the length of the crypts to the mucosal surface and that the proteinase was released apically. Indeed, by in situ hybridization, we detected matrilysin mRNA in epithelial cells at the base of crypts in the regenerative epithelium in the middle and proximal colon (Fig. 4 , D' and E'). Matrilysin mRNA was often detected in epithelial cells at the base of the crypts, whereas its protein was seen in epithelial cells further up the crypts (Fig. 4D and 4E) . This pattern suggests that matrilysin was expressed in cells at the base and was turned off in daughter cells that moved up along the length of the crypts onto the mucosal surface.
Altered neutrophil influx in Mmp7–/– mice
Matrilysin was expressed during periods of injury, lethality, and repair, and we hypothesized that increased mortality and more rapid onset of injury in wild-type mice resulted from exacerbation of acute inflammation. Few yet equal numbers of neutrophils were present in lamina propria of control (Day 0) wild-type and Mmp7–/– mice (data not shown). At Day 4 post-DSS, neutrophil influx was evident in wild-type mice, with pronounced accumulation in the middle colon near the base of crypts in the mucosa and submucosa of Grades 1 and 2 lesions (Fig. 5
). At this stage, fewer neutrophils had infiltrated into Mmp7–/– colons.
At Day 7, prominent neutrophil influx was evident in the middle colon of all mice. Although total MPO levels indicated that the number of infiltrated neutrophils was similar in tissue extracts between wild-type and Mmp7–/– mice at all times post-DSS (see Fig. 7 E), we saw a clear distinction between genotypes, where these leukocytes accumulated. In wild-type mice, most neutrophils had migrated into the mucosal layer, but in Mmp7–/– mice, neutrophils did not advance beyond the submucosa (Fig. 5 , D7'). In several areas of the middle colon of Mmp7–/– mice, neutrophils were densely packed in the submucosa just below the muscularis mucosa. At later times, the distribution and accumulation of neutrophils did not differ between wild-type and Mmp7–/– mice, and their numbers began to wane by Day 18 (Fig. 6 ). Consistent with the pattern of MMP12 mRNA, macrophage infiltration increased from Days 1 to 14 and decreased thereafter (data not shown). However, unlike neutrophils, we did not observe any difference in the location of macrophages between genotypes.
 View larger version (30K): [in a new window] |
Figure 7. Altered neutrophil migration and chemokine compartmentalization in Mmp7–/– mice. Whole colon homogenates (Tissue) and colon lavage were collected at various times after the onset of DSS exposure. Chemokine levels were quantified using a Luminex assay (A, B) or by ELISA (C, D). (E, F) MPO levels were determined by ELISA (n=4–6 per point). MPO levels differed significantly between genotypes at Day 11 (P=0.0420) but not at the other times. KC and MIP levels differed significantly between genotypes at Day 7 (P=0.0387 and 0.0188, respectively) but not at the other times.
|
Consistent with the immunostaining observations, MPO levels in colon tissue increased progressively over the first 2 weeks after onset of DSS, dropped thereafter, and did not differ between genotypes (Fig. 7E ). The temporal increase and drop in MPO levels correlated with the changes seen in matrilysin expression (Fig. 4C) . In contrast, MPO levels detected in colon lavage differed markedly between genotypes (Fig. 7F) . In wild-type mice, MPO levels in colon lavage peaked at Day 11, at which time, the influx of neutrophils into the tissue and matrilysin expression were increasing, and dropped thereafter. However, in Mmp7–/– mice, little MPO was detected in colon lavage at any time. Although the levels of total MPO measured in the wild-type colon lumens were much less than the levels measured in tissue, we cannot use these numbers to accurately determine what percent of neutrophils advanced into the lumen. In colon homogenates, MPO levels reflect the number of neutrophils that had accumulated in the tissue; however, in the colon lavage samples, MPO levels indicate the number of neutrophils present in the lumen at the time of sacrifice and that have not yet been eliminated. Regardless of these considerations, the MPO data (Fig. 7E and 7F) combined with the immunostaining observations (Fig. 5) indicate that the transmucosal migration of neutrophils was retarded in Mmp7–/– mice.
Chemokine expression and compartmentalization
To explore the cause for the retarded advancement of neutrophils in Mmp7–/– mice, we assessed the levels of various chemokines and inflammatory cytokines in tissue homogenates and colon lavage samples. Overall, there was no significant difference in the chemokine/cytokine levels in colon homogenates between wild-type and Mmp7–/– mice at any time post-DSS. For example, KC and MIP-2, acute-phase CXC chemokines and potent neutrophil chemotactic factors, were induced in response to DSS, although KC levels peaked predictably sooner than did MIP-2 and returned to basal levels during the reparative phase (Fig. 7A
and 7C)
. Similarly, the levels of IFN-, IL-1β, IL-6, IL-10, IL-12/p70, CCL2/MCP-1, and TNF- increased rapidly, reaching a peak by Day 7, and decreased thereafter toward baseline in both genotypes (data not shown). There was little change in IL-4 levels.
To assess if a defect in chemokine gradient formation contributed to the retarded transepithelial advancement of neutrophils that we saw in Mmp7–/– colons, we assessed the levels of KC and MIP-2 in intestinal lavage fluid and feces. Baseline levels of both CXC chemokines were detected in Day 4 colon lavage of both genotypes (Fig. 7B and 7D) . Coincident with the peak levels detected in tissue homogenates (Fig. 7A and 7C) , relatively high levels of KC and MIP-2 protein were measured in lumenal lavages collected at Day 7 from wild-type mice, mirroring the levels detected in tissue, and dropped thereafter. In contrast, the levels of KC and MIP-2 in colon lavage from Mmp7–/– mice remained at baseline values at all times examined (Fig. 7B and 7D) . These data suggest that matrilysin controls neutrophil movement by generating transmucosal gradients of CXC chemokines. As the peak in chemokine levels measured in colon lavage preceded the peak in MPO levels, the amount of KC and MIP-2 detected did not reflect a contribution from neutrophils.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In addition to impaired re-epithelialization, we observed impaired neutrophil infiltration in an inflamed Mmp7–/– colon. Although essentially equivalent numbers (based on total MPO levels) of neutrophils had effluxed from the vasculature into injured colons, their further advancement was retarded in Mmp7–/– mice. Whereas neutrophils were dispersed throughout the mucosa/submucosa in wild-type mice, they were largely confined to the submucosa (in which resides the vasculature bed from which they emigrate) in Mmp7–/– mice. As neutrophils do not express matrilysin [27 , 29 ], the defect in neutrophil migration into and across the mucosal layer was a consequence of a lack of expression by epithelial cells. We saw no overt difference in the influx of macrophages between the genotypes, indicating that matrilysin functions to specifically control neutrophil influx. Similarly, in acute lung injury, the influx of neutrophils, but not of macrophages, is impaired in Mmp7–/– mice [29 ].
The difference in DSS-mediated lethality was quite striking between genotypes. Recently, another group, using mice from our colony, reported that Mmp7–/– mice are more susceptible to DSS-induced mortality than wild-type mice [52
]. Although seemingly opposed to our findings, we attribute this discrepancy to the relatively small number of mice (n=5 per genotype) of unstated gender used in their study. Our lethality findings are based on data from eight experiments incorporating over 100 mice per genotype, of which data from six of these reflect the overall population trend shown in Figure 1D
. In two experiments, incorporating nine and 15 wild-type males and eight and 16 Mmp7–/– males, respectively, we observed no significant difference between genotypes (males only) in percent survival, which was overall 
30%. In these two experiments, there was a large difference in survival between wild-type (5%) and Mmp7–/– (74%) female mice. In another four experiments, essentially all wild-type male mice (94–100%) were dead within 12 days. Furthermore, we noted a significant gender bias in susceptibly to DSS-induced injury, which was consistently seen in all but one experiment, in which 90% of the wild-type females died (n=10 at Day 0) versus 60% of the wild-type males (n=15). Thus, in addition to sample size, gender differences may have skewed the findings reported by Shi et al. [52
].
We conclude that the resistance of the Mmp7–/– mice was not related to re-epithelialization, which was impaired in the knockout mice. Overall, the extent of osmotic injury caused by DSS consumption did not differ between phenotypes. Indeed, our observations demonstrate that the onset and initial degree of colon injury were similar between genotypes. However, once injury had begun, the wild-type tissue was apparently more effective at recruiting neutrophils and guiding these cells into and beyond the mucosal layer than the Mmp7–/– colon. Thus, matrilysin functions as part of an epithelial mechanism that permits neutrophil advancement across a wounded mucosal surface and their subsequent activation.
The impaired transepithelial advancement of neutrophils suggests that matrilysin also controls the activity or compartmentalization of chemokines. In mice, acute neutrophil influx is guided from the vasculature and into and through tissue by the CXC chemokines KC and MIP-2 [53 ]. Although we saw no difference in the levels of these chemokines in colon tissue homogenates between genotypes, we did see markedly reduced levels of both factors in colon lavages of Mmp7–/– mice. Similarly, in a model of lung injury, matrilysin generated a transepithelial gradient of KC, but in the lung, we saw no difference in the distribution of MIP-2 [29 ], suggesting that the mechanism of matrilysin activity may not be the same among tissues. In the lung injury model, we determined that KC accumulates on the glycosaminoglycan side-chains of syndecan-1, a transmembrane heparan sulfate proteoglycan, and that this complex is shed from the surface of epithelial cells by matrilysin, thereby establishing a chemokine gradient. Furthermore, in the lung injury, we reported that shedding of E-cadherin is dependent on matrilysin and speculated that proteolysis of this junctional protein is required for re-epithelialization [28 ]. However, in the colon model, we found no difference in shedding of syndecan-1 or E-cadherin in vivo or in tissue explants (data not shown), indicating that other proteinases function in release of membrane molecules in the intestine. Thus, although matrilysin affects similar processes among injured tissues (i.e., re-epithelialization, chemokine gradient formation, and transepithelial influx and activation of neutrophils), the mechanisms seem distinct.
Determining precisely how matrilysin controls neutrophil influx in the colon may point to a mechanism that governs acute inflammation, and our ongoing studies are designed to identifying the important substrate(s). We do not propose, however, that matrilysin controls the only mechanism involved in the transepithelial migration of neutrophils. For example, hepoxilin A3, an epithelial-derived eicosanoid, is required for neutrophils to cross an epithelial barrier in response to bacterial infection [54 , 55 ]. We do not yet know, however, if the production or activity of this lipid mediator is influenced by matrilysin. We predict that in the absence of matrilysin, neutrophils are stopped at the epithelial/interstitial interface, possibly by ligation to a cell-bound substrate of the proteinase. Thus, the matrilysin-controlled mechanism may function as a checkpoint to prevent premature neutrophil activation.
The observation that Mmp7–/– mice are protected against acute injury and lethality does not mean that this MMP is a "bad" proteinase. Indeed, matrilysin serves beneficial functions in immunity, epithelial migration, and neutrophil influx. Neutrophils comprise an essential cellular component of innate immunity, particularly in defense against bacteria. However, with extensive injury, as produced in many experimental models, a massive neutrophil influx would cause indiscriminate, severe, and potentially mortal damage. We have described what we believe to be a normal and required repair process, which if unregulated and exuberant, can be harmful. A lack of neutrophil activation and in turn, the oxidative damage it can cause, would be consistent with the protective effect we observed in the Mmp7–/– mice.
In summary, our studies show that matrilysin functions in re-epithelialization and more importantly, the transepithelial movement of neutrophils in injured colon and apparently, by mechanisms distinct from those in other tissues. These findings demonstrate that MMPs can serve beneficial functions, but they also indicate that MMPs can contribute to disease pathogenesis at the same time and place. Clearly, this family of proteinases did not expand during evolution to cause disease but rather to serve needed functions, which typically involve processes related to immunity and wound repair—in other words, re-establishment of tissue homeostasis. However, when overexpressed, MMPs can cause exuberant responses leading to tissue damage and disease.
Received January 9, 2008; revised February 13, 2008; accepted February 14, 2008.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
receptors are essential in a mouse model of transfusion-related acute lung injury J. Clin. Invest. 116,1615-1623[CrossRef][Medline]-defensin activation by the metalloproteinase matrilysin in innate host defense Science 286,113-117This article has been cited by other articles:
 |
|
 |
|
  A. M. Manicone, I. Huizar, and J. K. McGuire Matrilysin (Matrix Metalloproteinase-7) Regulates Anti-Inflammatory and Antifibrotic Pulmonary Dendritic Cells That Express CD103 ({alpha}E{beta}7-Integrin) Am. J. Pathol., December 1, 2009; 175(6): 2319 - 2331. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
