To the Editor, Mast cells are implicated in asthma, and an emerging body of evidence suggests that proteases released by mast cells are crucial players in asthma pathogenesis.1 The mast cell–restricted proteases, that is tryptase, chymase and carboxypeptidase A3 (CPA), are stored in large amounts as active enzymes in secretory granules in mast cells. Experimental studies in mouse models suggest that tryptase can contribute to airway hyperreactivity (AHR), a hallmark of asthma, whereas chymase can attenuate several features of asthma including AHR and IL-33 accumulation in lungs.2, 3 In contrast, there is very limited information on the role of CPA3 in asthma pathogenesis. However, findings from clinical studies suggest a link between Th2-high asthma and expression of CPA3 in the lungs.4, 5 CPA3 is a proteolytic enzyme with CPA-like activity, which means that it can remove aromatic and aliphatic residues at the C-terminal end of target proteins and peptides.6 Considering that lung mast cells release large amounts of proteolytically active CPA3 upon degranulation, we hypothesized that CPA activity in lung tissue could have a profound effect on the asthma response. Here, we investigated the contribution of CPA activity to features of asthma using the Nerita versicolor carboxypeptidase inhibitor (NvCI) in a mouse model of allergic asthma. NvCI is the strongest natural CPA inhibitor described so far, with inhibitory constants in the picomolar range, and also displays remarkable bioavailability and resistance.7 NvCI is highly specific for the M14A carboxypeptidase subfamily, including CPA3. To confirm the inhibitory effect of NvCI on mast cell CPA3, we used a chromogenic CPA substrate (M-2245/AAFP). We found that addition of NvCI completely inhibited CPA activity in supernatants of murine mast cells that were degranulated by IgE cross-linking or calcium ionophore (Figure 1A). Moreover, we detected weak CPA activity in human lung homogenates that was undetectable after adding NvCI (Figure 1B). These findings confirm that NvCI is a potent inhibitor of CPA activity derived from mast cells and located in lung tissue. We then investigated the effect of CPA inhibition in a house dust mite (HDM)–induced asthma model. Female mice of the BALB/c strain were pretreated with intraperitoneal injections of NvCI followed by intranasal instillations with HDM extract twice weekly for three weeks (Figure S1). Control groups of mice received HDM, buffer (PBS) or NvCI alone, at the same days. Resident mast cells in mouse lungs contain substantial amounts of CPA,1, 8 and we found increased degranulation of these mast cells upon HDM exposure (Figure S2), hence indirectly confirming the release of CPA in this model. Moreover, we confirmed the absence of CPA-like activity in the HDM extract (data not shown) and found that NvCI treatment of mice reduced the total CPA activity in lung homogenates (Figure 1C). We assessed AHR by measuring the change in lung resistance and dynamic compliance in response to nebulized methacholine compared with baseline values (PBS controls). As expected, mice challenged with HDM exhibited increased lung resistance and decreased dynamic compliance compared with control mice, whereas mice injected with NvCI alone did not differ from controls (Figure 1D). Notably, there was no AHR in the group of NvCI-treated HDM-challenged mice, as demonstrated by normalization of airway resistance and dynamic compliance in these mice (Figure 1D). This strongly supports the notion that CPA activity contributes to the development of AHR in the HDM-induced asthma model. To determine whether the NvCI-mediated attenuation of AHR was accompanied by a decrease in airway inflammation, we analysed HDM-induced accumulation of total cells, eosinophils, macrophages, lymphocytes and neutrophils in bronchoalveolar lavage (BAL). We found that NvCI treatment did not have any statistically significant effect on total or differential cell counts in BAL, although there was a trend for decreased numbers of eosinophils in NvCI-pretreated, HDM-challenged mice (Figure 1E-I). To determine the influence of NvCI on tissue inflammation and remodelling in the asthma model, lung sections were stained with periodic acid-Schiff (PAS) or haematoxylin and eosin and analysed for goblet cell accumulation, inflammatory cell infiltrates and thickening of airway smooth muscle layer. The results showed that NvCI significantly decreased accumulation of goblet cells in airway epithelium (Figure 2A, B), suggesting that CPA activity contributes to epithelial remodelling induced by HDM exposure. In agreement with the lack of effect on the inflammatory cells of the BAL fluid, NvCI did not influence the HDM-induced infiltration of inflammatory cells around airways or blood vessels (Figure 2C-E). Thickness of the smooth muscle layer in lung tissue did not differ between groups, and hence, no influence of NvCI was detected (Figure 2F). We have previously shown that extracellular levels of IL-13 and IL-33 can be limited by proteolytic degradation exerted by mast cell serine proteases, including chymase.3, 9 Notably, intra-epithelial mast cells in asthma exhibit an aberrant protease profile (tryptase+, CPA3+ and chymase−)4 and was recently shown to augment epithelial IL-33,10 suggesting an overall detrimental role of these mast cells. To examine the effect of CPA inhibition on cytokine concentrations, we measured a panel of pro-inflammatory cytokines in lung tissue extracts by ELISA. Concentrations of IL-5, IL-33 and CXCL-1 were increased upon HDM challenge, but there was no statistical difference between NvCI-treated and non-treated groups of HDM-challenged mice (Figure 2G). Hence, we found no evidence for a regulatory role of CPA activity on levels of cytokines in this model. To summarize, we demonstrated that NvCI can efficiently inhibit CPA activity derived from murine mast cells or human lungs. In the HDM-induced asthma model, pre-treatment with NvCI reduced the lung CPA activity and protected against AHR and reduced goblet cell hyperplasia, whereas inflammatory cell infiltration and cytokine levels were largely unaffected. This suggests that CPA activity is critical for AHR development and contributes to epithelial remodelling in this asthma model. It also suggests that CPA activity has a selective, local effect on airway smooth muscle and epithelium, compatible with the demonstrated accumulation of mast cells in these tissue compartments in asthma.1, 4 Although inhibition of mast cell CPA3 is a likely explanation for the dampening effect of NvCI in the asthma model, inhibition of other M14A carboxypeptidases cannot be excluded.7 To discriminate between the role of CPA3 and other carboxypeptidases, future studies including a battery of selective inhibitors of such enzymes and knockout mice are warranted. Our findings provide significant evidence for a detrimental role of carboxypeptidase in experimental asthma and highlight the biomedical potential of targeting carboxypeptidase activity in future asthma therapies. We thank Dr. Gunnar Pejler, Uppsala University, for his valuable scientific advice and comments on our data, and Drs. María Ángeles Chávez, Maday Alonso-del-Rivero and colleagues, Centre for Protein Research, Faculty of Biology, Universidad de La Habana, for help in productive collaborative research on marine protease inhibitors. The authors declare that they have no conflicts of interest. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.