Abstract
Mechanical ventilation is a life-saving therapy that can also damage the lungs. Ventilator-induced lung injury (VILI) promotes inflammation and up-regulates matrix metalloproteinases (MMPs). Among these enzymes, MMP-8 is involved in the onset of inflammation by processing different immune mediators. To clarify the role of MMP-8 in a model of VILI and their relevance as a therapeutic target, we ventilated wild-type and MMP-8–deficient mice with low or high pressures for 2 hours. There were no significant differences after low-pressure ventilation between wild-type and knockout animals. However, lack of MMP-8 results in better gas exchange, decreased lung edema and permeability, and diminished histological injury after high-pressure ventilation. Mmp8−/− mice had a different immune response to injurious ventilation, with decreased neutrophilic infiltration, lower levels of IFN-γ and chemokines (LPS-induced CXC chemokine, macrophage inflammatory protein–2), and significant increases in anti-inflammatory cytokines (IL-4, IL-10) in lung tissue and bronchoalveolar lavage fluid. There were no differences in MMP-2, MMP-9, or tissue inhibitor of metalloproteinase–1 between wild-type and knockout mice. These results were confirmed by showing a similar protective effect in wild-type mice treated with a selective MMP-8 inhibitor. We conclude that MMP-8 promotes acute inflammation after ventilation with high pressures, and its short-term inhibition could be a therapeutic goal to limit VILI.
Matrix metalloproteinase (MMP)–8 promotes acute lung inflammation. Genetic ablation or pharmacologic inhibition of this enzyme leads to an attenuated inflammatory response that ameliorates ventilator-induced lung injury. Our results point to MMP-8 inhibition as a therapeutic target to avoid acute inflammation after high-pressure ventilation.
Individual MMPs may have opposite effects. Nonselective inhibition of MMPs results in decreased lung injury after high-pressure ventilation due to a reduction in the inflammatory response (4, 12). However, these results preclude any conclusion about the precise enzymes involved in this beneficial effect. Moreover, different MMPs may have opposite effects. We have previously shown that absence of MMP-9 worsens VILI (9), and other authors have shown a similar protective role of MMP-9 in different models of lung injury (9, 13–16). These data suggest that other metalloproteinases should be responsible for the benefits of MMP inhibition in lung injury. MMP-8 (also known as collagenase-2 or neutrophil collagenase) deserves special attention, as this enzyme is synthesized by neutrophils, stored in secretory granules, and released during the inflammatory response. MMP-8 can cleave native collagen, and is essential for in vivo chemokine processing (17, 18). Critically, in a model of acute liver injury (19), mice lacking MMP-8 have decreased neutrophil infiltration, resulting in reduced tissue damage.
Accordingly, in this study, we examined whether MMP-8 is a mediator of VILI, and, consist with this, whether its absence correlates to a reduction of inflammation and lung injury after high-pressure ventilation. To test this hypothesis, we compared the effects of different ventilatory strategies in genetically modified mice lacking MMP-8 and their wild-type counterparts, studying the differences in gas exchange, lung edema, histological injury, and inflammatory response. Finally, the effects of a selective MMP-8 inhibitor were studied to explore a putative clinical application of our findings.
Some of these results have been previously reported in the form of an abstract (20).
All the experiments were performed according to the guidelines of the Committee on Animal Experimentation of the Universidad de Oviedo (Oviedo, Spain).
Mice used were 10- to 12-week-old males lacking MMP-8 (Mmp8−/−), of a mixed C57BL/6J/129 background (17), and wild-type littermates, and were kept in specific pathogen–free conditions.
These mice have no baseline differences in lung collagen content, MMP-2, MMP-9, or in any of the cytokines studied (see the online supplement).
Wild-type and knockout mice were anesthetized (ketamin and xylazin, intraperitoneally), and a tracheostomy performed. Briefly, a 20-gauge catheter was inserted in the trachea and tightened to avoid air leaks, and the animals were then ventilated (Evita 2 Dura with Neoflow; Dräger, Lübeck, Germany) in pressure-controlled mode. Mice were randomly assigned to one of two ventilatory strategies: low pressure (peak inspiratory pressure [PIP], 15 cm H2O [PIP 15]; positive end-expiratory pressure [PEEP], 2 cm H2O; respiratory rate, 100 breaths/min) or high-pressure ventilation (PIP, 25 cm H2O [PIP 25]; PEEP, 0 cm H2O; respiratory rate, 50 breaths/min). These pressure settings result in tidal volumes around 10 and 25 ml/kg at the start of the experiment, and respiratory rates were selected to achieve normocapnia, both based on preliminary experiments. Inspiratory-to-expiratory ratio was set at 1:1. Inspiratory and expiratory pauses were done after the onset of ventilation to exclude leaks or air trapping. Different experimental models of VILI lead to different responses in gas exchange, lung permeability, and tissue injury (12, 21, 22), depending on the ventilatory settings, time, and mouse strain (23). Our model results in a highly reproducible injury after 2 hours of ventilation (9, 24), with the main characteristics of lung dysfunction, including impairment in oxygenation, an increase in lung permeability, and structural damage.
After 2 hours of ventilation, a laparotomy was performed, and the aorta punctured to draw an arterial blood sample for gases measurement (NPT7; Radiometer Medical, Brønshøj, Denmark). The animal was then killed by exsanguination, the chest open, and the heart–lung block dissected and removed. The right lower lobe was weighed, dried in an oven (50°C for 48 h), and weighed again to calculate the wet-to-dry weight ratio. The remaining right lung was homogenized in a lysis buffer containing 20 mM Tris, 300 mM sucrose, 1% Triton X-100, and a protease inhibitor cocktail without EDTA (Complete; Roche, Berlin, Germany) and stored at −80°C.
Animals of both genotypes were subjected to the ventilatory strategies described previously here. A bronchoalveolar lavage (BAL) was performed at the end of the ventilatory period, before animals were killed. Four aliquots (1 ml) of saline were injected through the tracheostomy tube and recovered to isolate BAL fluid (BALF). Cell count and populations were measured through the use of a Neubauer chamber and in a hemocytometer. The remaining BALF was immediately frozen at −80°C. No other samples were collected from these animals. The protein content of the lysates and BALF was quantified (BCA protein assay; Pierce, Thermo Scientific, Rockford, IL).
The left lung was fixed with intratracheal formaldehyde at 20 cm H2O and immersed in the same fixative for 24 hours. After fixation, the left lung was embedded in paraffin and processed for a standard hematoxylin and eosin staining. Three slices, with a minimum separating distance of 1 mm, were scored by a pathologist blinded to the genotype and experimental conditions. A semiquantitative scale was used, scoring congestion and edema, hemorrhage, inflammatory cells, and septal thickening (scored from 0 to 4 each) (25).
Levels of MMP-8 in lung homogenates and BALF were determined by Western blotting. Samples containing 50 μg of protein or 12 μl of BALF were loaded in an 8% SDS–polyacrylamide gel. Gels were run and proteins transferred to a nitrocellulose membrane. These membranes were blocked with 3% nonfat milk, and incubated with a polyclonal antibody against murine MMP-8 raised in rabbits. A secondary peroxidase-linked antibody (Anti-Rabbit IgG; Cell Signaling, Danvers, MA) was added and bands of MMP-8 were detected by chemiluminiscence (Immobilon; Millipore, Billerica, MA). Membranes were scanned in a LAS-3000 mini camera (Fujifilm, Barcelona, Spain) and intensity of bands measured (in arbitrary density units) using ImageJ software (National Institutes of Health, Bethesda, MD). Purified murine MMP-8 from bone marrow cell culture supernatant was used as positive control. Tissue inhibitor of metalloproteinase (TIMP)–1 in lung tissue was measured by Western blotting following the same protocol, but using 15% SDS-polyacrylamide gels and a polyclonal antibody purchased from Abcam (Cambridge, UK). Recombinant TIMP-1 was used as positive control.
MMP-9 is stored in neutrophils together with MMP-8. Previous studies in models of chronic inflammation have shown that MMP-8−/− mice have a compensatory increase in MMP-9 (26, 27). To address this issue, we measured the activity of MMP-2 and MMP-9 in lung tissue or BALF using gelatin zymography. The method is described in detail in the online supplement. Briefly, the volume of lung homogenate corresponding to 95 μg of protein or 12 μl of BALF were loaded in an 8% SDS-polyacrylamide gel containing 0.2% gelatin, electrophoresed, and incubated overnight in a buffer. After staining with Coomassie blue and destaining, gelatinolytic activity appears as white bands in a blue background. The gels were then scanned, and the intensity of the bands quantified with ImageJ software (National Institutes of Health).
IL-1β, IFN-γ, macrophage inflammatory protein (MIP)–2, LPS-induced CXC chemokine (LIX), IL-4, and IL-10 were quantified in lung homogenates (with a volume corresponding to 50 μg of protein) and BALF (8 μl) with standard ELISA kits purchased from BD Biosciences (IL-4, IL-10; Madrid, Spain) or R&D Systems (the remaining molecules; Minneapolis, MN).
Myeloperoxidase was quantified as a marker of neutrophil infiltration in BALF. BALF (50 μl) were incubated with O-dianisidine and H2O2 in phosphate buffer, as previously described (28), and light absorbance at 460 nm was measured.
Total lung collagen was measured in homogenized lungs with the Sircol assay (Biocolor, Carrickfergus, UK), following the manufacturer's instructions. Briefly, lung tissue homogenates were mixed with the Sircol dye and centrifuged at 12,000 rpm for 15 minutes. The pellet was resuspended in NaOH, and the optical density at 540 nm was measured.
Wild-type mice were treated with five intraperitoneal doses of 40 mg/kg of the cyclohexylamine salt of (R)-1-(3′methylbiphenyl-4-sulfonylamino)-methylpropyl phosphonic acid, a specific MMP-8 inhibitor with no activity against other MMPs or a disintegrin and a metalloproteinase (ADAM)s (29), with 12-hour intervals between doses. Wild-type littermates were treated with the same volume of vehicle (200 μl of 5% DMSO in PBS). At 1 hour after the last dose, mice were anesthetized and ventilated with the high-pressure protocol. After 2 hours of ventilation, an arterial blood sample was drawn here to measure blood gases, and lung was harvested and processed as previously described here to measure wet-to-dry weight ratio, histological injury, MMP-2 and -9 activity, and IL-4, IL-10 and IFN-γ levels.
Data are expressed as means (±SD). Normal distribution of the data was assessed with a Kolgomorov-Smirnov test. The effects of genotype and ventilatory strategy were compared with a two-way ANOVA. Post hoc tests were done, when appropriate, using the Bonferroni correction. The differences between vehicle- and inhibitor-treated mice were compared with a t test. A P value of less than 0.05 was considered significant. All calculations were done with SPSS 15.0 software (SPSS Inc., Chicago, IL).
A total of 76 animals were included in the study, with 10 wild-type and 10 knockout mice assigned to each ventilatory group in the main experiment. BAL was performed in five additional animals from each genotype and ventilatory group. Finally, eight wild-type mice were treated with either the specific MMP-8 inhibitor or vehicle, and ventilated with high pressures to assess the efficacy of MMP-8 inhibition in the prevention of VILI. All of the animals survived the ventilatory protocol.
After low-pressure ventilation, the levels of MMP-8 in wild-type animals were low in lung tissue and undetectable in BALF (Figure 1). However, ventilation with high pressures resulted in a significant increase in MMP-8 in both lung parenchyma and BALF (Figures 1A and 1B). As expected, no MMP-8 was detected in Mmp8−/− animals (Figures 1C and 1D).
Figure 1. Matrix metalloproteinase (MMP)–8 increases in lung tissue homogenates (A) (n = 10/group) and bronchoalveolar lavage fluid (BALF) (B) (n = 5/group) after high-pressure ventilation. Representative Western blots are presented of (C) lung tissue and (D) BALF, showing no detectable MMP-8 in knockout mice. #P < 0.05 versus peak inspiratory pressure (PIP) 15 in post hoc tests.
[More] [Minimize]To investigate whether the increase in MMP-8 produced after high-pressure ventilation could contribute to VILI, mice lacking MMP-8 were ventilated simultaneously to their wild-type counterparts to compare the effects of mechanical ventilation between genotypes. No differences in gas exchange, histological lung injury, or wet-to-dry weight ratio between wild-type and knockout mice were observed after low-pressure ventilation (Table 1). Conversely, after high-pressure ventilation, there was a significant impairment in arterial oxygenation in Mmp8+/+ mice, with no changes in arterial carbon dioxide pressure (PaCO2) or arterial pH. This alteration observed in wild-type mice was related to an increase in the histological score of lung injury and in lung edema, quantified using the wet-to-dry weight ratio. Lungs from these animals showed an increase in septal edema, a prominent inflammatory infiltrate, and perivascular hemorrhages (Figure 2C). However, mice lacking MMP-8 developed only moderate lung injury after high-pressure ventilation, with no significant changes in arterial oxygen pressure:fraction of inspired oxygen ratio, in the histological scores or in lung edema quantification (P = 0.16, 0.13, and 0.08 in post hoc tests, respectively; Table 1). A mild septal edema and a reduced inflammatory infiltrate were the only findings in the histological study of these animals (Figure 2D), indicating a role of MMP-8 in VILI.
|
|
PIP 15 |
PIP 25 |
||||
|---|---|---|---|---|---|---|
|
|
Mmp8+/+
|
Mmp8−/−
|
Mmp8+/+
|
Mmp8−/−
|
||
| PaO2/FiO2, mm Hg | 408 ± 59 | 386 ± 46 | 290 ± 49# | 345 ± 60* | ||
| PaCO2, mm Hg | 52 ± 13 | 47 ± 14 | 40 ± 5 | 42 ± 15 | ||
| pH | 7.31 ± 0.09 | 7.34 ± 0.10 | 7.31 ± 0.25 | 7.39 ± 0.14 | ||
| HCO3, mEq/L | 24.6 ± 1.9 | 23.3 ± 2.8 | 20.1 ± 8.1 | 23.9 ± 4.3 | ||
| Wet-to dry weight ratio | 4.10 ± 0.75 | 4.48 ± 0.24 | 6.00 ± 0.69# | 5.12 ± 0.96* | ||
| Histological score
|
0.67 ± 0.82
|
0.75 ± 0.53
|
3.1 ± 1.24#
|
1.44 ± 0.73*
|
||
Figure 2. Representative slides stained with hematoxylin-eosin from wild-type and knockout mice after ventilation with low (PIP 15) or high (PIP 25) pressures. Note the mild septal thickening in knockout mice after injurious ventilation compared with the increased injury in wild-type mice.
[More] [Minimize]Parallel studies revealed the absence of significant differences between Mmp8+/+ and Mmp8−/− mice in BALF protein content or cell count after low-pressure ventilation (Figures 3A and 3B), although myeloperoxidase activity was significantly lower in knockout mice (Figure 3C). After ventilation at high pressures, wild-type animals showed a significant increase in BALF protein content, cell count, and myeloperoxidase activity, suggesting increased alveolocapillary permeability and polymorphonuclear infiltration. In contrast, Mmp8−/− mice ventilated at high pressures showed smaller amounts of protein and no significant changes in cell count or myeloperoxidase activity (Figure 3). In spite of the differences in absolute cell counts, there were no differences between wild-type and knockout mice in percentages of neutrophils in peripheral blood (wild-type, 9 ± 5.75%; knockout, 14 ± 6.6%; n = 4; P = 0.28) or BALF (wild-type, 28.6 ± 13.2%; knockout, 33.6 ± 14.6%; P = 0.72), suggesting that the relative composition of the cellular infiltrates is not influenced by MMP-8.
Figure 3. Markers of lung injury in BALF (n = 5/group). Differences in protein content (A), cell count (B), and myeloperoxidase activity (C) between wild-type and MMP-8−/− mice show a decreased protein and cell influx in the airspace in the latter. *P < 0.05 versus wild-type in post hoc tests; #P < 0.05 versus PIP 15 in post hoc tests.
[More] [Minimize]To determine if the effect of MMP-8 on VILI was related to its collagenolytic activity, total collagen was measured in homogenized lungs from animals subjected to both ventilator strategies. This analysis showed that ventilation using high pressures increased total lung collagen, although no differences were found between wild-type (26.6 ± 4.7 versus 31.2 ± 3.6 μg/mg protein for PIP 15 and PIP 25, respectively; P < 0.05) and knockout mice (23.7 ± 3.4 versus 33.6 ± 3.9 μg/mg protein for PIP 15 and PIP 25, respectively; P < 0.05), suggesting that the effect of MMP-8 on VILI is not dependent on its collagenolytic activity.
MMP-8 has been shown to process different cytokines and chemokines in vivo, affecting the inflammatory response. On this basis, we investigated whether the different response to ventilation between wild-type and Mmp8-deficient mice could be due to changes in the immune response. For this purpose, we determined the concentration of T helper (Th) 1 (IL-1β and IFN-γ) and Th2 (IL-4 and IL-10) cytokines, as well as two CXC chemokines (MIP-2 and LIX) in lung homogenates and BALF (Figure 4). The inflammatory response caused by high-pressure ventilation resulted in a significant up-regulation of these six mediators in lung tissue when compared with low-pressure ventilation, and all except IL-4 and MIP-2 were also up-regulated in BALF. Mmp-8–deficient mice showed a blunted response in the increases of MIP-2 in lung tissue and LIX in BALF. These mutant animals also showed significantly lower levels of IFN-γ in lung tissue and BALF, irrespective of the ventilatory strategy used. Moreover, high-pressure ventilation resulted in higher levels of Th2 cytokines (IL-4 and IL-10) only in Mmp8−/− mice subjected to injurious ventilation.
Figure 4. Cytokine (IL-1β, IFN-γ, IL-4, IL-10) and chemokine (macrophage inflammatory protein [MIP]–2, LPS-induced CXC chemokine [LIX]) levels in lung homogenate (n = 10/group) and BALF (n = 5/group) for each experimental group (see main text for details). *P < 0.05 versus wild-type in post hoc tests; #P < 0.05 versus PIP 15 in post hoc tests.
[More] [Minimize]To investigate whether the absence of MMP-8 could be compensated by other MMPs present in the lung, we measured the activities of MMP-2 and MMP-9 in lung homogenates and BALF by gelatin zymography. After high-pressure ventilation, there was an increase in MMP-9 activity in lung homogenates and BALF, whereas MMP-2 increased only in the latter (Figures 5A–5D). However, there were no differences between the genotypes in the activities of MMP-2 or MMP-9. There were no significant differences in TIMP-1 between ventilatory strategies or genotypes (Figure 5E). Figure 5F shows representative zymography and Western blots of these measurements. Together, these results indicate that neither gelatinases (MMP-2 and MMP-9) nor TIMP-1 compensate for the loss of MMP-8 in mice.
Figure 5. Absence of MMP-8 does not trigger a compensatory gelatinolytic increment. Gelatinase activity ([A and B] MMP-9; [C and D] MMP-2) was measured in lung homogenates (n = 10/group) and BALF (n = 5/group). In the same sense, there were no differences in tissue inhibitor of metalloproteinase (TIMP)–1 in lung tissue (E). Representative zymographs and Western blotting (F) are shown. #P < 0.05 versus PIP 15 in post hoc tests.
[More] [Minimize]The data presented in this study suggest that MMP-8 activity contributes to the tissue damage caused by injurious ventilation. Therefore, to test the clinical relevance of the findings obtained with Mmp8−/− mice, we evaluated the effect of an MMP-8–selective inhibitor in VILI. Accordingly, wild-type mice were treated either with the cyclohexylamine salt of (R)-1-(3′methylbiphenyl-4-sulfonylamino)-methylpropyl phosphonic acid or with vehicle (DMSO). No mice showed any sign of toxicity caused by the drug or the vehicle, and all survived the ventilatory period. Mice treated with the inhibitor showed a better arterial oxygenation than vehicle-treated mice after 2 hours of high-pressure ventilation (Figure 6A). In agreement with these findings, we observed that animals treated with the MMP-8–specific inhibitor had lower wet-to-dry weight ratios and histological scores of lung injury (Figures 6B–6E). In addition, there was a trend toward an increase in PaCO2 in vehicle-treated mice (60.3 ± 25.2 versus 40.6 ± 14.1 mm Hg; P = 0.10), with no significant differences in pH (7.20 ± 0.12 versus 7.30 ± 0.15; P = 0.16). No signs of atelectasis were documented in the lungs of these animals, although the increase in PaCO2 could suggest the existence of areas of hypoventilation or increased alveolar dead space in these extensively injured lungs.
Figure 6. Effects of MMP-8 inhibition in wild-type mice. Oxygenation (A), lung wet-to-dry weight ratio (B), and histologic injury (C) were measured in wild-type mice treated with an MMP-8–specific inhibitor (n = 8) or only with vehicle (n = 8), showing the benefits of MMP-8 inhibition in ventilator-induced lung injury. Representative slides (hematoxylin and eosin staining) of vehicle (D) and inhibitor-treated (E) mice are shown. *P < 0.05 versus vehicle-treated mice.
[More] [Minimize]The effects of inhibitory therapy on selected cytokines and MMPs were studied (Figure 7). Treatment with the inhibitor was associated with an increase in IL-10 in lung tissue homogenates, with no differences in IL-4 or IFN-γ. To eliminate the possibility of compensatory changes in MMP-2 and/or MMP-9, gelatin zymography was performed. There were no differences in the activity of these enzymes between inhibitor- or vehicle-treated mice. Collectively, these results demonstrate that selective pharmacological inhibition of MMP-8 has a positive effect on VILI in this experimental model.
Figure 7. Treatment with the specific MMP-8 inhibitor does not modify MMP-2 or MMP-9 activities (A–C), IFN-γ (D), or IL-4 (E) in lung tissue, but significantly increases the anti-inflammatory cytokine, IL-10 (F) (n = 8/group). *P < 0.05 versus vehicle-treated mice.
[More] [Minimize]
Our results demonstrate that MMP-8 plays a key role in VILI by promoting a proinflammatory response within the lung, but not by contributing to changing the collagen content of the extracellular matrix. Results from Mmp-8 knockout mice show that the absence of this enzyme shifts the inflammatory response to an anti-inflammatory one, decreasing tissue damage and improving lung function. The relevance of our findings is reinforced by the demonstration of decreased VILI after treatment with an MMP-8–selective inhibitor. Importantly, these results suggest that MMP-8 inhibition could be an effective therapeutic approach to treating this type of lung injury.
MMP-8 is stored in neutrophils and released at the sites of inflammation (26). Its substrates are native collagen, as well as a number of immune mediators, including several cyto- and chemokines. It can inactivate IL-10, and splenocytes from Mmp8−/− mice release more IL-4 and IL-10 after a stimulus (27). Other studies have also emphasized the relevance of the proteolytic activity of MMP-8 over CXC chemokines. Processing of LIX (a murine ortholog of IL-8) by MMP-8 increases its chemotactic activity (30), whereas Mmp8-null mice have a delayed peak in MIP-2 after injury (26). By these mechanisms, MMP-8 modulates the onset and the clearance of inflammation. Thus, mice lacking this enzyme have a delayed neutrophilic infiltration after an inflammatory stimulus, but also a slower clearance of it (17, 26, 30). These features make these mutant mice more resistant to acute liver injury (19) or autoimmune encephalomyelitis (27).
MMP-8 increases in patients with both acute and chronic lung injuries, such as the acute respiratory distress syndrome (8), chronic obstructive pulmonary disease (31), and pulmonary fibrosis (32). Furthermore, our data show a highly significant increase of this MMP after injurious ventilation, consistent with recent studies on MMP-8 (33) and the increases observed for other MMPs (4, 10, 34). More interesting is the finding of decreased VILI in MMP-8–deficient animals after 2 hours of ventilation. These benefits were not related to changes in collagen content of the lung. The small but statistically significant increase in collagen content of the lungs is consistent with the results of other groups, which found an increase in procollagen gene expression after VILI (35). Of note, no signs of fibrosis were found in the histological study; there were no differences between genotypes in this result. It is possible that collagen turnover is regulated by overlapping mechanisms, with MMP-13 being the main collagenolytic enzyme in mice.
However, we did find significant differences in the inflammatory response; specifically, a decrease in CXC chemokines (MIP-2 and LIX) and an increase in anti-inflammatory cytokines (IL-4 and IL-10) in mutant mice. Importantly, treatment with a specific MMP-8 inhibitor increased IL-10 levels. These results suggest that MMP-8 could be an important regulator of IL-10 function in vivo. Similar benefits of an anti-inflammatory response mediated by IL-10 during VILI have been previously shown (24, 36–38). Finally, we have observed lower levels of IFN-γ in mutant animals, irrespective of the ventilatory strategy, reinforcing the idea of a modified inflammatory response in Mmp8−/− mice. However, this cytokine did not change with inhibitor therapy, so its pathogenetic relevance is less clear.
Our results differ from those recently reported by Dolinay and colleagues (33). They have found that Mmp8−/− mice develop more alveolar permeability than their wild-type counterparts after 8 hours of mechanical ventilation with moderate tidal volumes (10 ml/kg) and PEEP of 2 cm H2O, with no differences in lung edema or cell/neutrophil count in BALF. The comparison with our work suggests that inflammation and permeability in VILI could be independent phenomena with different time patterns, and with MMP-8 acting as a modulator of both processes. For instance, lack of MMP-8 is associated with a slower clearance of the inflammatory infiltrate in a model of skin wounds, causing an impaired healing (26). Similarly, such a persistent inflammation within the lungs could result in chronic damage. These results highlight that the benefits of MMP-8 inhibition should be viewed with caution in regard to long-term therapy. In contrast, long-term MMP-8 inhibition improved outcome in a model of autoimmune encephalomyelitis (27).
In addition to the uncertainties related to long-term inhibition discussed previously here, our methodology has some limitations that should be discussed:
Animal model: the injurious ventilatory strategy tested has no clinical counterpart. However, findings from other studies using similar strategies have shown similarities to those in patients under high-volume ventilation in regard to inflammatory response (39).
Hemodynamic factors: the role of vascular injury in VILI has been widely studied, and hemodynamic factors could be relevant. These were not measured in our study, and, therefore, a role of MMP-8 in these mechanisms of VILI cannot be discarded. However, the absence of metabolic acidosis at the end of the experiment allows us to discard a severe systemic hypoperfusion.
Inhibitor therapy: dosages and timing of the phosphonate were empirically chosen based on published results with other inhibitors. The impact of other doses or administration schedules has not been determined. Moreover, lung injury in DMSO-treated mice was increased when compared with nontreated, wild-type mice. It has been described that DMSO can aggravate lung injury (40, 41), but the solvent was unavoidable due to the lipophilic nature of the substance. This does not invalidate our results, as both vehicle- and inhibitor-treated mice received equal amounts of DMSO.
Avoidance of lung injury during mechanical ventilation is an important goal in both patients with and those without lung injury. However, there are patients at risk of ventilator-associated lung injury even during protective ventilation (42). Moreover, limitation of one mechanism of VILI, such as alveolar instability with high PEEP values, can lead to alveolar overstretching (43), complicating the determination of an optimal equilibrium when setting the ventilator. For these reasons, the combination of a protective ventilatory strategy and a drug aimed to limit VILI could be a promising therapeutic strategy. Previous groups (4, 12) have demonstrated a decrease in VILI with broad-spectrum MMP inhibitors (with activity against MMP-1, -2, -9, and other enzymes, such as inducible nitric oxide synthase or cyclooxygenase-2 [44]). However, data from models using knockout mice suggest that some MMPs could play a protective role. Specifically, the absence of MMP-9 is related to a more severe injury after high-pressure ventilation (9), abdominal sepsis (13), ozone exposure (14), or in bronchopulmonary dysplasia (15). Hence, the search for specific MMPs related to tissue injury is warranted to develop specific inhibitors that could have a clinical application. Unfortunately, although some drugs have shown beneficial effects in experimental models, none has been tested in clinical trials.
The results observed in mice deficient for MMP-8 activity, rendered by both genetic manipulation and chemical inhibition, illustrate the potential of this enzyme as a therapeutic target in acute VILI. According to our observations, targeting MMP-8 could be useful during periods of high-pressure ventilation, such as recruitment maneuvers or in patients with the most severe disease. However, the benefits seen in our study, and the concerns over the safety of MMP-8 inhibition at later time points found by other researchers, suggest that this strategy should be viewed with caution. Therefore, more complex experiments, using two-hit models, for example, should be assessed before any clinical recommendation can be provided.
Our results show that MMP-8 plays a key role in triggering a proinflammatory response in the lungs after injurious ventilation. Furthermore, we show that absence or inhibition of this enzyme results in decreased lung injury, probably by enhancing the anti-inflammatory response. Therefore, MMP-8 inhibition could be a useful strategy to prevent VILI. However, slower clearance of the inflammatory infiltrates seen in other models of injury using Mmp8−/− animals warrants more studies aimed at clarifying the safety of prolonged therapy against MMP-8.
The authors thank M.S. Pitiot, A.R. Folgueras, C. Garabaya, and A. Lopez for their help with the histological and biochemical work, and A. Ramsay for his careful review of the manuscript.
| 1. | Slutsky AS. Ventilator-induced lung injury: from barotrauma to biotrauma. Respir Care 2005;50:646–659. |
| 2. | Gattinoni L, Carlesso E, Cadringher P, Valenza F, Vagginelli F, Chiumello D. Physical and biological triggers of ventilator-induced lung injury and its prevention. Eur Respir J Suppl 2003;47:15s–25s. |
| 3. | Kheradmand F, Werner E, Tremble P, Symons M, Werb Z. Role of rac1 and oxygen radicals in collagenase-1 expression induced by cell shape change. Science 1998;280:898–902. |
| 4. | Foda HD, Rollo EE, Drews M, Conner C, Appelt K, Shalinsky DR, Zucker S. Ventilator-induced lung injury upregulates and activates gelatinases and emmprin: attenuation by the synthetic matrix metalloproteinase inhibitor, prinomastat (ag3340). Am J Respir Cell Mol Biol 2001;25:717–724. |
| 5. | Nagase H, Visse R, Murphy G. Structure and function of matrix metalloproteinases and TIMPs. Cardiovasc Res 2006;69:562–573. |
| 6. | Demoule A, Decailliot F, Jonson B, Christov C, Maitre B, Touqui L, Brochard L, Delclaux C. Relationship between pressure–volume curve and markers for collagen turn-over in early acute respiratory distress syndrome. Intensive Care Med 2006;32:413–420. |
| 7. | Ricou B, Nicod L, Lacraz S, Welgus HG, Suter PM, Dayer JM. Matrix metalloproteinases and TIMP in acute respiratory distress syndrome. Am J Respir Crit Care Med 1996;154:346–352. |
| 8. | Fligiel SE, Standiford T, Fligiel HM, Tashkin D, Strieter RM, Warner RL, Johnson KJ, Varani J. Matrix metalloproteinases and matrix metalloproteinase inhibitors in acute lung injury. Hum Pathol 2006;37:422–430. |
| 9. | Albaiceta GM, Gutierrez-Fernandez A, Parra D, Astudillo A, Garcia-Prieto E, Taboada F, Fueyo A. Lack of matrix metalloproteinasec9 worsens ventilator-induced lung injury. Am J Physiol Lung Cell Mol Physiol 2008;294:L535–L543. |
| 10. | Haseneen NA, Vaday GG, Zucker S, Foda HD. Mechanical stretch induces MMP-2 release and activation in lung endothelium: role of emmprin. Am J Physiol Lung Cell Mol Physiol 2003;284:L541–L547. |
| 11. | Greenlee KJ, Werb Z, Kheradmand F. Matrix metalloproteinases in lung: multiple, multifarious, and multifaceted. Physiol Rev 2007;87:69–98. |
| 12. | Kim JH, Suk MH, Yoon DW, Lee SH, Hur GY, Jung KH, Jeong HC, Lee SY, Suh IB, Shin C, et al. Inhibition of matrix metalloproteinasec9 prevents neutrophilic inflammation in ventilator-induced lung injury. Am J Physiol Lung Cell Mol Physiol 2006;291:L580–L587. |
| 13. | Renckens R, Roelofs JJ, Florquin S, de Vos AF, Lijnen HR, van't Veer C, van der Poll T. Matrix metalloproteinasec9 deficiency impairs host defense against abdominal sepsis. J Immunol 2006;176:3735–3741. |
| 14. | Yoon HK, Cho HY, Kleeberger SR. Protective role of matrix metalloproteinasec9 in ozone-induced airway inflammation. Environ Health Perspect 2007;115:1557–1563. |
| 15. | Lukkarinen H, Hogmalm A, Lappalainen U, Bry K. Matrix metalloproteinasec9 deficiency worsens lung injury in a model of bronchopulmonary dysplasia. Am J Respir Cell Mol Biol 2009;41:59–68. |
| 16. | Cabrera S, Gaxiola M, Arreola JL, Ramirez R, Jara P, D'Armiento J, Richards T, Selman M, Pardo A. Overexpression of MMP9 in macrophages attenuates pulmonary fibrosis induced by bleomycin. Int J Biochem Cell Biol 2007;39:2324–2338. |
| 17. | Balbin M, Fueyo A, Tester AM, Pendas AM, Pitiot AS, Astudillo A, Overall CM, Shapiro SD, Lopez-Otin C. Loss of collagenase-2 confers increased skin tumor susceptibility to male mice. Nat Genet 2003;35:252–257. |
| 18. | Van Lint P, Libert C. Matrix metalloproteinase–8: cleavage can be decisive. Cytokine Growth Factor Rev 2006;17:217–223. |
| 19. | Van Lint P, Wielockx B, Puimege L, Noel A, Lopez-Otin C, Libert C. Resistance of collagenase-2 (matrix metalloproteinase–8)–deficient mice to TNF-induced lethal hepatitis. J Immunol 2005;175:7642–7649. |
| 20. | Albaiceta GM, Parra D, Garcia-Prieto E, Astudillo A, Fueyo A, Puente XS, Taboada F, Lopez-Otin C. Decreased ventilator-induced lung injury in matrix metalloproteinase–8 deficient mice. Intensive Care Med 2008;34:S96. |
| 21. | Kaynar AM, Houghton AM, Lum EH, Pitt BR, Shapiro SD. Neutrophil elastase is needed for neutrophil emigration into lungs in ventilator-induced lung injury. Am J Respir Cell Mol Biol 2008;39:53–60. |
| 22. | Wilson MR, Choudhury S, Takata M. Pulmonary inflammation induced by high-stretch ventilation is mediated by tumor necrosis factor signaling in mice. Am J Physiol Lung Cell Mol Physiol 2005;288:L599–L607. |
| 23. | Eckle T, Fullbier L, Grenz A, Eltzschig HK. Usefulness of pressure-controlled ventilation at high inspiratory pressures to induce acute lung injury in mice. Am J Physiol Lung Cell Mol Physiol 2008;295:L718–L724. |
| 24. | Pedreira PR, Garcia-Prieto E, Parra D, Astudillo A, Diaz E, Taboada F, Albaiceta GM. Effects of melatonin in an experimental model of ventilator-induced lung injury. Am J Physiol Lung Cell Mol Physiol 2008;295:L820–L827. |
| 25. | Belperio JA, Keane MP, Burdick MD, Londhe V, Xue YY, Li K, Phillips RJ, Strieter RM. Critical role for CXCR2 and CXCR2 ligands during the pathogenesis of ventilator-induced lung injury. J Clin Invest 2002;110:1703–1716. |
| 26. | Gutierrez-Fernandez A, Inada M, Balbin M, Fueyo A, Pitiot AS, Astudillo A, Hirose K, Hirata M, Shapiro SD, Noel A, et al. Increased inflammation delays wound healing in mice deficient in collagenase-2 (MMP-8). FASEB J 2007;21:2580–2591. |
| 27. | Folgueras AR, Fueyo A, Garcia-Suarez O, Cox J, Astudillo A, Tortorella P, Campestre C, Gutierrez-Fernandez A, Fanjul-Fernandez M, Pennington CJ, et al. Collagenase-2 deficiency or inhibition impairs experimental autoimmune encephalomyelitis in mice. J Biol Chem 2008;283:9465–9474. |
| 28. | Bradley PP, Priebat DA, Christensen RD, Rothstein G. Measurement of cutaneous inflammation: estimation of neutrophil content with an enzyme marker. J Invest Dermatol 1982;78:206–209. |
| 29. | Biasone A, Tortorella P, Campestre C, Agamennone M, Preziuso S, Chiappini M, Nuti E, Carelli P, Rossello A, Mazza F, et al. Alpha-biphenylsulfonylamino 2-methylpropyl phosphonates: enantioselective synthesis and selective inhibition of MMPs. Bioorg Med Chem 2007;15:791–799. |
| 30. | Tester AM, Cox JH, Connor AR, Starr AE, Dean RA, Puente XS, Lopez-Otin C, Overall CM. LPS responsiveness and neutrophil chemotaxis in vivo require PMN MMP-8 activity. PLoS One 2007;2:e312. |
| 31. | Vernooy JH, Lindeman JH, Jacobs JA, Hanemaaijer R, Wouters EF. Increased activity of matrix metalloproteinase–8 and matrix metalloproteinase–9 in induced sputum from patients with COPD. Chest 2004;126:1802–1810. |
| 32. | Manoury B, Nenan S, Guenon I, Lagente V, Boichot E. Influence of early neutrophil depletion on MMPs/TIMP-1 balance in bleomycin-induced lung fibrosis. Int Immunopharmacol 2007;7:900–911. |
| 33. | Dolinay T, Wu W, Kaminski N, Ifedigbo E, Kaynar AM, Szilasi M, Watkins SC, Ryter SW, Hoetzel A, Choi AM. Mitogen-activated protein kinases regulate susceptibility to ventilator-induced lung injury. PLoS One 2008;3:e1601. |
| 34. | Hasaneen NA, Zucker S, Cao J, Chiarelli C, Panettieri RA, Foda HD. Cyclic mechanical strain-induced proliferation and migration of human airway smooth muscle cells: role of emmprin and MMPs. FASEB J 2005;19:1507–1509. |
| 35. | de Carvalho ME, Dolhnikoff M, Meireles SI, Reis LF, Martins MA, Deheinzelin D. Effects of overinflation on procollagen type III expression in experimental acute lung injury. Crit Care 2007;11:R23. |
| 36. | Donnelly SC, Strieter RM, Reid PT, Kunkel SL, Burdick MD, Armstrong I, Mackenzie A, Haslett C. The association between mortality rates and decreased concentrations of interleukin-10 and interleukin-1 receptor antagonist in the lung fluids of patients with the adult respiratory distress syndrome. Ann Intern Med 1996;125:191–196. |
| 37. | Lee HS, Wang Y, Maciejewski BS, Esho K, Fulton C, Sharma S, Sanchez-Esteban J. Interleukin-10 protects cultured fetal rat type II epithelial cells from injury induced by mechanical stretch. Am J Physiol Lung Cell Mol Physiol 2008;294:L225–L232. |
| 38. | Hoegl S, Boost KA, Czerwonka H, Dolfen A, Scheiermann P, Muhl H, Zwissler B, Hofstetter C. Inhaled IL-10 reduces biotrauma and mortality in a model of ventilator-induced lung injury. Respir Med 2009;103:463–470. |
| 39. | Ranieri VM, Suter PM, Tortorella C, De Tullio R, Dayer JM, Brienza A, Bruno F, Slutsky AS. Effect of mechanical ventilation on inflammatory mediators in patients with acute respiratory distress syndrome: a randomized controlled trial. JAMA 1999;282:54–61. |
| 40. | Haschek WM, Baer KE, Rutherford JE. Effects of dimethyl sulfoxide (DMSO) on pulmonary fibrosis in rats and mice. Toxicology 1989;54:197–205. |
| 41. | Santos NC, Figueira-Coelho J, Martins-Silva J, Saldanha C. Multidisciplinary utilization of dimethyl sulfoxide: pharmacological, cellular, and molecular aspects. Biochem Pharmacol 2003;65:1035–1041. |
| 42. | Terragni PP, Rosboch G, Tealdi A, Corno E, Menaldo E, Davini O, Gandini G, Herrmann P, Mascia L, Quintel M, et al. Tidal hyperinflation during low tidal volume ventilation in acute respiratory distress syndrome. Am J Respir Crit Care Med 2007;175:160–166. |
| 43. | Albaiceta GM, Luyando LH, Parra D, Menendez R, Calvo J, Pedreira PR, Taboada F. Inspiratory vs. expiratory pressure–volume curves to set end-expiratory pressure in acute lung injury. Intensive Care Med 2005;31:1370–1378. |
| 44. | Steinberg J, Halter J, Schiller H, Gatto L, Carney D, Lee HM, Golub L, Nieman G. Chemically modified tetracycline prevents the development of septic shock and acute respiratory distress syndrome in a clinically applicable porcine model. Shock 2005;24:348–356. |
