German cockroach proteases and protease-activated receptor-2 regulate chemokine production and dendritic cell recruitment

Abstract

We recently showed that serine proteases in German cockroach (GC) feces (frass) decreased experimental asthma through the activation of protease-activated receptor (PAR)-2. Since dendritic cells (DCs) play an important role in the initiation of asthma, we queried the role of GC frass proteases in modulating CCL20 (chemokine C-C motif ligand 20) and granulocyte macrophage colony-stimulating factor (GM-CSF) production, factors that regulate pulmonary DCs. A single exposure to GC frass resulted in a rapid, but transient, increase in GM-CSF and a steady increase in CCL20 in the airways of mice. Instillation of protease-depleted GC frass or instillation of GC frass in PAR-2-deficient mice significantly decreased chemokine release. A specific PAR-2-activating peptide was also sufficient to induce CCL20 production. To directly assess the role of the GC frass protease in chemokine release, we enriched the protease from GC frass and confirmed that the protease was sufficient to induce both GM-CSF and CCL20 production in vivo. Primary airway epithelial cells produced both GM-CSF and CCL20 in a protease- and PAR-2-dependent manner. Finally, we show a decreased percentage of myeloid DCs in the lung following allergen exposure in PAR-2-deficient mice compared to wild-type mice. However, there was no difference in GC frass uptake. Our data indicate that, through the activation of PAR-2, allergen-derived proteases are sufficient to induce CCL20 and GM-CSF production in the airways. This leads to increased recruitment and/or differentiation of myeloid DC populations in the lungs and likely plays an important role in the initiation of allergic airway responses.

Fig. 1

A single instillation of GC frass induced CCL20 expression in the airways of mice. Balb/c mice were administered a single intratracheal instillation of PBS or GC frass (40 μg/40 μl) 1, 3, 6, and 18 h later. BAL fluid was harvested, clarified, and production of CCL20 (a) and GM-CSF (b) was analyzed by ELISA. Means ± SEM (n = 4-9 mice per group) are reported (∗ p < 0.001 compared to PBS control) from 3 separate experiments.

Fig. 2

GC frass proteases and PAR-2 regulate CCL20 expression in mice. a Balb/c mice were administered a single intratracheal instillation of PBS, aprotinin-treated PBS, GC frass (40 μg/40 μl) or protease-depleted frass. BAL fluid was harvested 18 h later and clarified, and CCL20 was analyzed by ELISA. Means ± SEM (n = 6 mice per group) are reported (∗ p = 0.001; ∗ ∗ p = 0.01) from a single experiment. b Wild-type or PAR-2-deficient mice were given a single intratracheal instillation of PBS or GC frass and CCL20 levels in BAL fluid were analyzed 18 h later. Means ± SEM (n = 8 mice per group) are reported (∗ p = 0.001; ∗ ∗ p = 0.021) from 2 separate experiments.

Fig. 3

GC frass proteases and PAR-2 regulate GM-CSF expression in mice. a Balb/c mice were administered a single intratracheal instillation of PBS, aprotinin-treated PBS, GC frass (40 μg/40 μl) or protease-depleted frass and harvested 3 h later. BAL fluid was harvested, clarified, and GM-CSF was analyzed by ELISA. Means ± SEM (n = 6 mice per group) are reported (∗ p < 0.001; ∗ ∗ p = 0.01) from 2 separate experiments. b Wild-type or PAR-2-deficient mice were given a single intratracheal instillation of PBS or GC frass and GM-CSF levels in BAL fluid were analyzed 18 h later. Means ± SEM (n = 8 mice per group) are reported (∗ p < 0.001; ∗ ∗ p = 0.021) from 2 separate experiments.

Fig. 4

Direct activation of PAR-2 with protease or PAR-2 agonist increased chemokine production. Naïve mice were administered a single instillation of protease-enriched GC frass (0.5 U) or GC frass (40 μg) and BAL fluid was harvested 3 or 18 h later. a GMCSF levels at 3 h after instillation (∗ p < 0.05). b CCL20 levels at 3 h after instillation (∗ p < 0.05). c CCL20 levels at 18 h after instillation (∗ p < 0.05). d Naïve mice were administered a single instillation of PBS, PAR-2-activating peptide (PAR-2-AP), or PAR-2 control peptide (PAR-2-CP) and analyzed 18 h after instillation (∗ p < 0.05). In all cases, means ± SEM (n = 4-8 mice per group) are reported from 1 or 2 separate experiments.

Fig. 5

CCL20 and GM-CSF expression from MTECs. Tracheas from mice were isolated and MTECs were cultured in transwell plates until confluent. MTECs were treated with PBS, aprotinintreated PBS, GC frass (300 ng/ml), or protease-depleted frass GC frass (300 ng/ml) for 18 h. Cell supernatants were harvested, clarified and analyzed for CCL20 (a; ∗ p < 0.001, ∗ ∗ p = 0.017) or GMCSF (b; ∗ p = 0.001, ∗ ∗ p = 0.005) by ELISA. Mean ± SEM are reported from 5 separate experiments. MTECs were cultured from wild-type or PAR-2-deficient mice. Cell supernatants were harvested, clarified and analyzed for CCL20 (c; ∗ p < 0.001, ∗ ∗ p = 0.019) or GM-CSF (d; ∗ p = 0.002, ∗ ∗ p = 0.004) by ELISA. Mean ± SEM are reported from 3 or 4 separate experiments.

Fig. 6

Percentage of mDCs and allergen uptake in mDCs following AF405-GC frass exposure. Wild-type and PAR-2-deficient mice were exposed to a single intratracheal instillation of PBS or Alexa-Fluor (AF) 405-labeled GC frass (40 μg/40 μl). Whole lungs were isolated 20 h later and cells were dissociated and stained for flow cytometric analysis. Cells were gated only on mDCs (CD11c+, CD11b+, Gr1-, CD317-). a Percentage of mDCs in the lung (∗ p < 0.001, ∗ ∗ p = 0.007). b Percentage of AF405-positive mDCs in the lung (∗ p < 0.001). Mean ± SEM are reported from 6 mice per group with the samples being run in quintuplicate in a single experiment.