COPD is a lung disease characterized by progressive airflow limitation due to the inflammation-driven destruction of alveolar walls. Neutrophils play a key role in the chronic inflammatory response in COPD. Since cigarette smoke exposure is the main risk factor for the development of COPD we investigated the effect of CSE on the breakdown of whole pulmonary matrix collagen into the chemotactic collagen fragment N-ac-PGP by human neutrophils.
We are the first to show that cigarette smoke can activate neutrophils to generate the chemotactic collagen fragment N-ac-PGP from whole collagen in vitro. Incubation of PMNs with CSE activated these cells, which resulted in the release of the chemo-attractant CXCL8 and the proteases MMP8 and MMP9. Simultaneous incubation of PMNs with CSE and collagen resulted in N-ac-PGP generation. In addition, PMNs constitutively expressed PE activity and protein. Simultaneous incubation of PMNs with the tripeptide N-ac-PGP resulted in the release of CXCL8, MMP8 and MMP9.
Moreover, we tested whether PMNs from COPD patients are different from PMNs from healthy donors. Although incubation of PMNs from COPD patients with different CSE concentrations tended to release more CXCL8, this did not reach the level of significance when compared to PMNs of healthy donors. Interestingly, here we show that the intracellular basal PE activity of PMNs from COPD patients is a 25-fold higher when compared to healthy donors. Immunohistological staining of human lung tissue specimens for PE protein showed that besides inflammatory cells, including neutrophils and macrophages also epithelial cells express significant levels of PE protein.
Early in inflammation, neutrophils migrate from the capillaries into the interstitial space, following a chemotactic gradient of CXCL8 
. At the site of inflammation neutrophils are activated, leading to the release of more CXCL8 
. This release leads to a self-perpetuating inflammatory state where neutrophils attract more neutrophils via chemokine receptors CXCR1 and CXCR2 
. Recently, we showed that cigarette smoke extract (CSE) can act as a chemo-attractant for PMNs 
. This led to the question whether CSE may activate the neutrophil to synthesize CXCL8, acting in an autocrine/paracrine fashion. shows that the activation of PMNs by CSE exposure leads to the release CXCL8. We hypothesize that once infiltrated in the lung tissue, cigarette smoke activates the infiltrated neutrophils. This activation results in a CXCL8 release by the neutrophils, which in turn will attract more neutrophils into the airways.
The increased expression of MMPs is considered to be a key factor in the development of COPD. In this study, the MMP8 and MMP9 release by PMNs was elevated after cigarette smoke and N-ac-PGP exposure to human neutrophils. These results are in accordance with clinical data from different groups. It was shown that although MMP8 and MMP9 levels are lower in smokers when compared to COPD patients 
, the MMP levels from both groups are elevated when compared to non-smokers 
. Here we show that CSE-stimulated COPD neutrophils did not produce more MMP-9 in comparison to the neutrophils of healthy donors (Figure S1
). However, it has been published that COPD patients have higher neutrophil counts in the bronchoalveolar lavage fluid 
. This leads to the conclusion that the increase in MMP9 levels in COPD patients is the result of an increase in neutrophil number and not due to an increase in MMP9 release.Besides MMP8 and MMP9, PE is needed to generate PGP from whole collagen; the MMPs cleave whole collagen in fragments of 30 to 100 amino acids in length, after which PE specifically cleaves PGP from these smaller fragments 
. Recently, it was published that neutrophils contain PE 
, which is confirmed in this study. PE activity was measured in lysates of PMNs. Incubation of PMNs with CSE or N-ac-PGP did not affect intracellular PE activity, which suggests that PE is constitutively active. Although PE activity could be measured in the supernatant of CSE or N-ac-PGP incubated PMNs, these levels were very low. We hypothesize that cigarette smoking causes a locally restricted lung inflammation where necrotic neutrophils or neutrophils undergoing NETosis release PE to the exterior, which contributes to PGP generation. This can be substantiated with data from and ; incubating PMNs for 16 hours with CSE resulted in a decrease in cell viability, PE release and subsequent generation of N-ac-PGP from whole collagen. It is possible that other cells besides neutrophils play a role in collagen destruction by supplying PE. shows that also pulmonary alveolar macrophages express PE. Neutrophils and macrophages present in lung tissue of current smokers and COPD patients with GOLD stage II and IV highly expressed PE, while the number of inflammatory cells and consequently the PE expression was decreased in the lung tissue of ex-smokers.
The next step was to investigate the effect of CSE on the breakdown of whole collagen into collagen fragment N-ac-PGP by human neutrophils. The multistep pathway of collagen breakdown has been studied in a murine model of cigarette smoke-induced lung emphysema in our group by Braber et al. 
. There it was demonstrated that all relevant components (neutrophils, MMP8, MMP9 and PE) involved in this pathway to generate (N-ac-)PGP from collagen were upregulated in the airways exposed to cigarette smoke, suggesting that activation of cells by cigarette smoke leads to the release of proteases and extracellular matrix breakdown. Although this murine model showed that (N-ac-)PGP is formed after cigarette smoke exposure in the airways, here we demonstrate using in vitro
techniques that upon stimulation with CSE the human neutrophil is able to breakdown collagen into N-ac-PGP fragments. Our results confirm the findings of O'Reilly et al
, who found that in vitro
human neutrophils were capable to generate PGP from whole collagen after LPS exposure 
Neutrophils contain all necessary components for PGP generation and in this report we demonstrated that simultaneous incubation of these cells with CSE and collagen leads to PGP generation. Although N-ac-PGP levels were measurable after a 16 hour incubation period, non-acetylated PGP could not be detected in these supernatants. This can be explained by the fact that cigarette smoke itself is responsible for N-terminally acetylating PGP, thereby enhancing its chemotactic capacity 
Louhelainen et al. 
and Miller et al. 
showed that smoking cessation improved lung function although elevated neutrophil counts and the protease burden in the airways continued for months. An explanation for the elevated neutrophil influx and protease levels after smoke cessation is that the continued neutrophil chemotaxis and activation is mediated via N-ac-PGP. In this study we demonstrate that this chemotactic tripeptide can activate neutrophils to release CXCL8 that will lead to an increase in neutrophilic migration. In addition, N-ac-PGP also induced the release of MMP8 and MMP9 from neutrophils, which will result in more collagen breakdown and formation of N-ac-PGP. It was recently published that N-ac-PGP can induce the release of MMP9, which is confirmed in this study. It was indicated that extracellular matrix-derived N-ac-PGP could result in a feed-forward cycle by releasing MMP-9 from activated PMNs through the ligation of CXCR1 and CXCR2 and subsequent activation of the ERK1/2 MAPK 
Tissue destruction is a hallmark of COPD. Since PE is essential in the collagen breakdown process, we measured the basal intracellular PE activity in PMNs from COPD patients. Interestingly, the basal PE activity of PMNs from COPD patients was remarkably higher than in PMNs from healthy donors, which suggests that PE can play an important role in lung collagen breakdown leading to the development of COPD. This phenomenon is not limited to COPD; Gaggar et al.
also observed that the PE activity was elevated in cystic fibrosis samples when compared with normal controls 
. Moreover, here we suggest that the PMNs from COPD patients are activated to a greater extend, since the CXCL8 levels released by these PMNs appeared to be higher than from PMNs from healthy controls.
Finally, to our knowledge, this is the first in vitro study that indicates that neutrophils activated by cigarette smoke can destruct collagen into N-ac-PGP and that this collagen fragment can activate neutrophils, which may lead in vivo to a self-propagating cycle of neutrophil infiltration, chronic inflammation and lung emphysema.
From this study but also from other studies on collagen breakdown in COPD mice models 
, we can conclude that different treatment interventions are possible to tackle the ongoing inflammation observed in COPD. MMP inhibitors might be a valuable drug target; in addition to suppressing the accelerated extracellular matrix turnover, the generation of chemotactic N-ac-PGP can be counteracted. Since we show in that the basal intracellular PE activity of neutrophils is a 25-fold higher in COPD neutrophils in comparison to the neutrophils from healthy donor PE inhibitors such as valproic acid (VPA) might be useful. This has already been tested in an animal model; a significant decreased neutrophil influx in the BAL fluid of smoke-exposed mice was observed after treatment with VPA 
. Also, treatment with peptide L-arginine-threonine-arginine (RTR), which binds to PGP sequences, led to a decrease in neutrophil migration in mice exposed to smoke 
In conclusion, MMP-, PE- or PGP-inhibitors can serve as an attractive therapeutic target and may open new avenues towards effective treatment of COPD.