Pollen Proteases Play Multiple Roles in Allergic Disorders

Allergic diseases are a major health concern worldwide. Pollens are important triggers for allergic rhinitis, conjunctivitis and asthma. Proteases released upon pollen grain hydration appear to play a major role in the typical immunological and inflammatory responses that occur in patients with allergic disorders. In this study, we aimed to identify specific proteolytic activity in a set of pollens with diverse allergenic potential. Diffusates from Chenopodium album, Plantago lanceolata and Eucalyptus globulus were added to a confluent monolayer of Calu-3 cells grown in an air-liquid interface system. We identified serine proteases and metalloproteinases in all pollen diffusates investigated. Proteases found in these pollen diffusates were shown to compromise the integrity of the lung epithelial barrier by disrupting transmembrane adhesion proteins E-cadherin, claudin-1 and Occludin, as well as, the cytosolic complex zonula occludens-1 (ZO-1) resulting in a time-dependent increase in transepithelial permeability. Tight junction disruption and increased transepithelial permeability facilitates allergen exposure to epithelial sub-layers contributing to the sensitization to a wide range of allergens. These pollen extracts also induced an increase in the release of interleukin 6 (IL-6) and interleukin 8 (IL-8) cytokines measured by flow cytometry possibly as a result of the activation of protease-activated receptors 2 (PAR-2).

For the identification of proteolytic activity in the pollen extracts, these were prepared in non-denaturing conditions in order to preserve the enzymatic activity, and subjected to an electrophoresis in a polyacrylamide gel containing 1 mg/ml gelatin. After the electrophoretical separation, SDS was removed using Triton X-100, incubated overnight at 37 ºC and in buffer pH 7.4. Finally staining with Coomassie Blue allowed the detection of enzymatic digestion spots that corresponded to non-colored bands ( Figure SI2). Figure SI2. Proteolytic profiles of pollen extracts obtain by zymography 12% polyacrylamide copolymerized with 1 mg/ml gelatin. 1) 3.5 µ g of the initial pollen extract from C. album. 2) 5 µ g of the initial pollen extract from C. album. 3) 8 µ g of the initial pollen extract from E. globulus. 4) 12 µ g of the initial pollen extract from E. globulus. 5) 2 µ g of the initial pollen extract from P. lanceolata. 6) 3 µ g of the initial pollen extract from P. lanceolata.
Additionally, pollen proteolytic activity was rapidly characterized through twodimensional zymography. For first dimension, concentrated pollen extracts samples were submitted to an isoelectric focusing, using pH 3-10 strips, were separation occurred accordingly to their pI. The second dimension, separation occurs accordingly to their molecular mass, on a gelatin zymography. Revelation of enzymatic digested spots was possible through Coomassie Blue staining ( Figure SI3). Pollen extracts samples were concentrated using centricon filter devices of 30kDa, in order to increase the total quantity of protein. This method intends to remove proteins of low molecular weight and concentrate the samples of pollen extracts in high molecular weight proteins, region normally comprised of the majority of proteases. Figure SI3. Proteolytic 2D-PAGE profile of the concentrated pollen extracts of C. album, E. globulus and P. lanceolata. The total quantity of each sample of concentrated pollen extract was ~100 µ g.
In conclusion, it was shown that all pollen diffusates contained gelatin degrading proteases which are of high molecular weight and acidic pI ( Figure SI1-3).

Effect of pollen diffusates on intercellular adhesion protein integrity
The effect of pollen diffusates on intercellular protein complexes was evaluated  Figure SI4). Noteworthy, is the fact that C. album was used in this assay at a 1:20 dilution as this pollen diffusate induced a high degree of cell detachment.  Figure SI4. Effect of pollen diffusates on intercellular adhesion protein integrity analyzed using an immunofluorescence assay. Calu-3 cells were incubated with pollen diffusates from Chenopodium album (0.013 ± 0.01 mg/ml), Plantago lanceolata (0.18 ± 0.03 mg/ml) and Eucalyptus globulus pollen diffusates (0.75 ± 0.18 mg/ml) for 6 h. The cells were also exposed to denatured pollen diffusates (95 ºC for 30 min). Cells incubated with culture medium were used as a control.
Representative images are shown for each stimulus. Image scale bar corresponds to 10 µ m.
Imaging to detect unspecific labelling was also performed and such labeling was shown to be non-existent (images not shown). (1:1000) and 543 mouse (1:1000) (Invitrogen)) for 1 h at room temperature. The fluorescent dye Hoechst 33342 (Sigma, 0.5 μg/ml) was used to stain nuclei. The transwell membrane inserts were cut and mounted on a fluorescent mounting medium (DAKO, Denmark A/S, Denmark) and imaging performed using an