Bac1 (RLCRIVVIRVCR), Bac5 (RFRPPIRRPPIRPPFYPPFRPPIRPPIFPPIRPPFRPPLGPFP), indolicidin (ILPWKWPWWPWRR), BMAP-27 (GRFKRFRKKFKKLFKKLSPVIPLLHLG), sBMAP-27 (GLPKLHFRRKKGKLVSFPLKFKFKIRL), BMAP-28 (GGLRSLGRKILRAWKKYGPIIVPIIRIG), and BMAP-34 (AGLFRRLRDSIRRGQQKILEKARRIGERIKDIFRG) were synthesized using solid-phase peptide synthesis by ChinaPeptides (Shanghai, China). All peptides showed >98% purity by HPLC analysis and mass spectrometry. Peptides were dissolved in 2.5% DMSO (Bac1, Bac5, indolicidin, BMAP-28, and BMAP-34) or deionized sterile water (BMAP-27 and sBMAP-27) and stored at −80°C. Polyinosinic–polycytidylic acid [poly(I:C)], and phorbol myristate acetate (PMA) were purchased from InvivoGen (Toulouse, France). Bovine rIFN-α was purchased from Kingfisher Biotech Inc., Saint Paul, MN, USA. Detection of reactive oxygen species (ROS) was performed with the dye CM-H₂DCFDA (Thermo Fisher Scientific, Switzerland). The integrity of the bacterial membrane was evaluated with BacLight™ Bacterial Membrane Potential Kit (Molecular Probes, Thermo Fisher Scientific).

Anti-human MxA (clone M143) was provided by Dr. J. Pavlovic (Institute of Medical Virology, University of Zürich, Zürich, Switzerland). Anti-mouse β-actin (clone AC-74) was obtained from Sigma-Aldrich (Buchs, Switzerland). Peroxidase-conjugated donkey anti-mouse IgG antibody was purchased from Jackson ImmunoResearch Labs (West Grove, PA, USA).

Bovine neutrophil granulocytes were isolated as previously described with a few adaptations (20, 21). Bovine blood sampling was performed in compliance with the Swiss animal protection law and approved by the animal welfare committee of the Canton of Bern, Switzerland, license number BE102/15. Briefly, 200 ml of blood was drawn from the jugular vein of cows at the Clinic for Ruminants (Vetsuisse Faculty, University of Bern, Switzerland) and was directly mixed with citrate-based anti-coagulant Alsever’s solution (1.55 mM of C₆H₁₂O₆, 408 mM of Na₃C₆H₅O₇⋅2H₂O, 1.078 mM of NaCl, and 43 mM of C₆H₈O₇, pH 6.2). Afterward, the blood was centrifuged at 1,000 × g at 4°C for 20 min, the plasma and Buffy coat layer aspirated, and erythrocytes lysed in 12 ml ice cold hypotonic lysing solution (pH 7.3, 10.56 mM of Na₂HPO₄, and 2.67 mM of NaH₂PO₄ in sterile water). After 90 s, 6 ml of ice cold hypertonic solution (pH 7.3, 10.56 mM of Na₂HPO₄, 2.67 mM of NaH₂PO₄, and 0.43 M of NaCl in sterile water) was added to restore isotonic conditions. The cells were centrifuged at 800 × g, 4°C for 5 min and erythrocyte lysis repeated 2–3 times until red blood cells were no longer visible. Cells were then pooled into two Falcon tubes with 30 ml ice cold PBS/EDTA. A solution of 10 ml Ficoll-Paque 1.084 g/ml (GE Healthcare Europe GmbH, Freiburg, Germany) was overlaid with the cell suspension and centrifuged at 400 × g at room temperature for 40 min without brake. Neutrophils were recovered in the pellet, washed once with PBS/EDTA, counted in Trypan blue solution, and resuspended at 5 × 10⁷ cells/ml in RPMI 1640 medium (Biochrom GmbH, Berlin, Germany). For each animal, experiments were run in duplicate cultures. A total of 1 ml of 5 × 10⁷ neutrophils was seeded in a 24-well plate and were either stimulated with 100 ng/ml of PMA or kept unstimulated in medium for 10 min. Cells were cultured at 37°C in humidified 5% CO₂ atmosphere. After 10 min, the plate was centrifuged at 400 × g and 4°C for 5 min. Supernatants were removed, and 1 ml medium was added to the cells. Neutrophils were incubated for a further 1 h at 37°C, before being centrifuged at 400 × g and 4°C for 5 min. Supernatants were collected and stored at −80°C until used. Aliquots of 5 × 10⁶ neutrophils per tube were used to quantify ROS. Briefly, neutrophils were incubated at 37°C with 5 µM of CM-H₂DCFDA in 100 µl of HBSS (Thermo Fisher Scientific) for 15 min. Afterward, cells were washed and incubated further in fresh medium for 30 min. Cells were washed one more time, suspended in 200 µl of Cell Wash^® (Becton Dickinson, Basel, Switzerland), and acquired on a FACSCantoII (Becton Dickinson).

All bacterial strains including S. aureus strains (M3842, M905-1, and M3850) (22), S. uberis strains (M100/11, BL246, and ALP8092) (23) and E. coli (9217/10) were isolated from bovine mastitis milk (24). All strains were maintained on Trypticase soy agar plates with 5% sheep blood (Becton Dickinson AG, Allschwil, Switzerland) and always freshly plated the day before the experiment. Bacterial colonies were suspended in 0.85% NaCl solution (Axon Lab AG, Baden, Switzerland) to reach an OD value of 0.5 (DensiCHECK PLUS, Biomérieux, Genève, Switzerland), corresponding to approximately 1–1.5 × 10⁸ cell/ml. The minimum inhibitory concentration (MIC) of cathelicidins was determined by the broth microdilution method as previously described (25). Bacteria were further diluted to approximately 1 × 10⁵ cells/ml in cation-adjusted Mueller-Hinton (MH) broth (Thermo Fisher Scientific). Several twofold peptide dilutions (40–0.3 µM) were prepared in a volume of 50 µl MH II broth (Becton Dickinson AG) in a round-bottom 96-well plate. Finally, 50 µl of 10⁵ cells/ml were added to the peptide dilutions to reach a final concentration of 5 × 10⁴ cells/ml. The MIC was determined as lowest concentration without visible bacterial growth after overnight incubation at 37°C.

To determine the antimicrobial effects of BMAP-27 or sBMAP-27, bacterial suspensions of strains M3842, M100/11, and 9217/10 were incubated with several twofold dilutions of the peptides in a final volume of 100 µl of MH medium. Briefly, a volume of 50 µl containing 10⁵ cells/ml was mixed with 50 µl peptide dilution and incubated at 37°C for 1 h. As control, bacteria were incubated in the absence of cathelicidin. Each sample was then diluted 10-fold in PBS, and 50 µl suspension was plated on MH II agar petri dishes (Becton Dickinson). Colonies were counted after overnight incubation at 37°C and reported as CFU/ml. The bactericidal activity of neutrophil supernatants was evaluated using a similar protocol. Briefly, 95 µl of supernatants were pipetted in a flat-bottom 96-well plate and inoculated with 5 µl of 10⁶ cells/ml of E. coli suspension. After incubation at 37°C for 1 h, the samples were diluted 10-fold in PBS and plated on MH agar petri dishes. Colonies were enumerated after overnight incubation at 37°C and were reported as CFU/ml.

To evaluate bacterial membrane potential, approximately 10⁷ bacterial cells were suspended in 1 ml of filtered PBS solution and treated with BMAP-27 or sBMAP-27. Carbonyl cyanide 3-chlorophenylhydrazone (CCCP) was employed at 5 µM as positive control. Afterward, E. coli and S. aureus were treated with 30 µM, while S. uberis was stained with 60 µM of DiOC₂(3). Cells were incubated 30 min at RT and acquired on a FACSCantoII. The gating strategy was defined to remove doublets and bacterial cell debris based on the different FSC and SSC parameters (Figure S1A in Supplementary Material). To evaluate the ability of cathelicidins to form pores in the membrane, around 10⁷ bacterial cells were suspended in 0.5 ml of filtered 0.85% NaCl solution in tubes and treated with BMAP-27 or sBMAP-27. As positive control, cells were incubated in 70% isopropanol. After 1 h incubation at RT, bacterial cell suspensions were centrifuged at 10,000 × g for 3 min and supernatants discarded. For staining of dead E. coli, 0.5 ml 7-aminoactinomycin D (7-AAD) at 1 µg/ml was used, while for S. aureus and S. uberis 0.5 ml of propidium iodide (PI) at 1 µg/ml was employed. Cells were incubated for 5 min at RT. Bacteria were further washed with 1 ml of 0.85% filtered NaCl, centrifuged at 10,000 × g for 3 min, and suspended in 0.5 ml of 0.85% filtered NaCl solution before being acquired on a FACSCantoII. Doublets, aggregates, and debris were excluded by electronic gating (Figure S1A in Supplementary Material).

Cathelicidins were diluted in 8 µl of nuclease-free water at the indicated concentration and mixed with 2 µl of 100 ng pEAK8-His plasmid. The plasmid alone was employed as a control. Preparations were incubated at RT for 10 min. Subsequently, 2 µl of 6× loading dye (Thermo Fisher Scientific) were added. Finally, the mixtures were run on a 0.8% agarose gel and DNA visualized with ethidium bromide under UV radiation.

Primary bovine turbinate epithelial cells (BT cells) were obtained as previously described (26) and the cultures maintained in DMEM supplemented with 10% FBS, non-essential amino acids (NEAA), neomycin/bacitracin, penicillin/streptomycin (stock solutions purchased from Biochrom), and GlutaMAX (Thermo Fisher Scientific). BT cells were plated in 24-well plates at 50,000 cells/well and incubated for 18 h at 37°C in 5% CO₂. For induction of Mx, cathelicidins were mixed with 100 ng/ml of poly(I:C) in a final volume of 1 ml, incubated for 10 min at room temperature, and then added to the cells. Alternatively, different dilutions of neutrophil-derived supernatants were mixed with poly(I:C) at 100 ng/ml in complete medium (DMEM with 10% FBS, NEAA, neomycin/bacitracin, penicillin/streptomycin, and GlutaMAX). Positive controls for Mx1 induction included cell treatment with 100 ng/ml of bovine rIFN-α and 1 µg/ml of poly(I:C). Cultures were further incubated at 37°C for 20 h. After removal of the medium, the cells were lyzed by addition of 30 µl M-PER mammalian protein extraction reagent (Thermo Fisher Scientific) supplemented with complete protease inhibitor cocktail (Roche Diagnostics, Basel, Switzerland). Lysates were centrifuged at 10,000 × g at 4°C for 10 min and 20 µl supernatant collected for western blotting. Separation and blotting of proteins were performed as described earlier (26).

For western blots, 20 µl of primary epithelial cell lysates, obtained from approximately 5–10 × 10⁴ cells, were loaded and separated on 12% SDS-polyacrylamide gels (Bio-Rad, Reinach, Switzerland). Proteins were then blotted onto nitrocellulose membranes (Amersham Biosciences, Dübendorf, Switzerland) followed by a saturation step with PBS-T buffer containing 0.5% Tween 20 (v/v) supplemented with 5% (w/v) low-fat dry milk (Nestlé, Vevey, Switzerland). Afterward, membranes were incubated with the anti-human MxA and anti-mouse β-actin antibodies in PBS-T buffer supplemented with 0.5% milk (w/v) at 4°C overnight. The next day, a washing step with PBS-T was performed for 30 min. Membranes were further incubated with the peroxidase-conjugated donkey anti-mouse IgG antibody in 0.5% milk, 0.5% Tween20 PBS for 1 h. Subsequently, membranes were washed twice in 0.5% Tween 20 in PBS. Finally, solutions of WesternBright ECL HRP substrate (Advansta Inc., Menlo Park, CA, USA) were mixed in a 1:1 ratio and incubated with membranes for 2 min. Images were obtained using a CCD-LAS3000 camera (Fuji Film).

For the detection of cathelicidins in the neutrophil supernatants, a rabbit anti-cathelin domain polyclonal antibody (18) was employed at 10 µg/ml. For detection, a peroxidase-conjugated donkey anti-rabbit IgG was used as secondary antibody (Thermo Fisher Scientific). Neutrophil lysates were employed as a positive control.

Helical wheel projection and net charge values were obtained using HeliQuest.¹ Representation of BMAP-27 and sBMAP-27 were generated using I-TASSER (27, 28). The best model for BMAP-27 obtained a C-score of −0.42, while sBMAP-27 only reached −2.28. Peptide conformations were also predicted on PsiPred² and Jpred.³ Analysis of flow cytometry files was undertaken using FlowJo v10. Data were analyzed and graphs generated using GraphPad Prism 6.0. Statistical analyses used matched paired t-test or multiple comparisons using one-way ANOVA. Dunnett’s multiple comparison post hoc test was employed to compare data with a control group. Significant differences are noted as *p < 0.05 or **p < 0.001.

A combination of chemical factors including hydrophobicity, net charge, and the ability of HDP to form particular secondary structures functionally determines the antimicrobial activity of peptides. Leaving the net charge of +10 intact, we designed a scrambled version of BMAP-27 (sBMAP-27) to evaluate the impact of structure on biological activity. The amphipathic conformation characteristic of alpha-helical cathelicidins was made visible in a helical wheel projection (Figure 1A), as well as in a three-dimensional model of the original BMAP-27 (Figure 1B). This contrasted with sBMAP-27, which was predicted to form a beta-sheet structure (Figure 1C).

As expected, sBMAP-27 had reduced antimicrobial activity when compared to BMAP-27. This was found with S. aureus, S. uberis, and E. coli and was in the range of a 4- to 10-fold reduction (Figure 2). These data suggest that the structural conformation of BMAP-27 mostly confers bactericidal function, although the role of the cationic charge cannot be excluded. Due to their amphipathic characteristics, cationic HDP typically affect the potential of bacterial membranes, which can be evaluated by flow cytometry using carbocyanine dyes (29). This test demonstrated that compared to sBMAP-27, BMAP-27 strongly affected membrane integrity of S. aureus (Figure 3A). Cathelicidin-induced changes of membrane potential were also observed with S. uberis and E. coli, but the impact of secondary structure was less evident (Figure 3B). In order to evaluate if BMAP-27 kills bacterial cells through the formation of pores (30), cell-impermeable nucleic acid stains were employed. For both Gram-positive bacteria, the frequency of PI-positive cells was always higher in the presence of BMAP-27 compared to its scrambled version (Figure 3C; Figure S1B in Supplementary Material). Similarly, E. coli treated with BMAP-27 showed a higher frequency of 7-AAD positive cells in comparison to sBMAP-27-treated E. coli (Figure 3D). Altogether, these data indicate that the bactericidal property of BMAP-27 is at least in part dependent on the amphipathic conformation, presumably impacting membrane integrity by forming pores into both Gram-negative and Gram-positive bacteria.

To evaluate the antimicrobial activity of bovine cathelicidins (5–10), six synthetic peptides including the alpha-helical BMAP-27, -28, and -34, the scrambled sBMAP-27, the tryptophan-rich indolicidin, the cysteine-rich disulfide-bridged peptide Bac1, and a proline-rich peptide Bac5, were employed to assess MIC against seven mastitis isolates (Table 1). Cathelicidins belonging to the alpha-helical peptide class were the most efficient in inhibiting bacterial growth. Surprisingly, indolicidin, Bac1 (dodecapeptide), and Bac5 had a low capacity to prevent the growth of all strains of S. aureus and E. coli at relevant in vivo concentrations (<20 µM). As reported above, 2- to 4-fold higher concentration of sBMAP-27 was required to reach MIC compared to BMAP-27, confirming the importance of the alpha-helical conformation to sustain optimal antimicrobial function.

Previous studies demonstrated that many cationic HDP including cathelicidins (19, 31–33) and β-defensins (34) could interact with negatively charged nucleic acids. Consequently, we determined the ability of bovine HDP to influence nucleic acid migration in agarose gels. Migration patterns were strongly altered at 10 µM of BMAP-27, sBAMP-27, BMAP-28, BMAP-34, and Bac5 (Figure 4). These observations were confirmed by quantifying the signal of the lower band on the gels (Figure S2 in Supplementary Material).

Both human and porcine cathelicidins have been described as forming complexes with free nucleic acids, favoring uptake by plasmacytoid dendritic cells (pDC), and leading to potent type I interferon (IFN) responses (19, 31). Considering that bovine pDC are very rare in peripheral blood (35), and that cathelicidins will also be secreted at mucosal surfaces, we employed a primary epithelial cell (BT cells) model and determined Mx1 induction as a readout for type I interferon responses (26). First, the titration of poly(I:C) promoting Mx1 expression was performed to evaluate the threshold at which these cells sensed poly(I:C). The induction of Mx1 was not detectable below a concentration of 156 ng/ml of poly(I:C) (Figure S3 in Supplementary Material). Consequently, 0.1 µg/ml was used to evaluate the ability of cathelicidins to promote nucleic acid sensing. Our results demonstrate that the addition of BMAP-27, sBAMP-27, BMAP-28, and BMAP-34 to poly(I:C) enabled the detection of Mx1, whereas this response was very weak or not observed with indolicidin, Bac1, or Bac5 (Figure 5A). Furthermore, a cathelicidin dose-dependent induction of Mx1 was shown with BMAP-27, sBAMP-27, BMAP-28, BMAP-34 as well as with the disulfide-bridged peptide Bac1 (Figure 5B). These data demonstrate that bovine alpha-helical and cysteine-rich disulfide-bridged cathelicidins can promote nucleic acid sensing at low peptide concentrations.

In order to clarify the importance and relevance of neutrophil-derived cathelicidins in mediating these responses, bovine blood neutrophils were used as a source of cathelicidins. To induce neutrophil activation, we employed PMA stimulation. The observed reduction in side scatter determined by flow cytometry, as well as the morphological changes observed by bright field microscopy (Figure 6A) indicated neutrophil degranulation following PMA stimulation. Activation was also confirmed by the quantification of ROS by flow cytometry (Figure 6B). Although some variations were noticed between animals, the fluorescence intensity of CM-H₂DCFDA (ROS indicator) of PMA-stimulated neutrophils was significantly higher compared to unstimulated neutrophils (Figure 6C). As expected, the activation of PMA-treated neutrophils was also characterized by the release of cathelicidins detected by western blot (Figures 6D,E). To evaluate antimicrobial function, neutrophil supernatants were tested for their ability to reduce the growth of E. coli and S. aureus. No significant effects were found with supernatants of unstimulated neutrophils when compared to medium with both E. coli and S. aureus. However, 3.7-fold E. coli and a 4-fold S. aureus CFU reduction was observed with supernatants from PMA-stimulated neutrophils compared to medium (Figures 7A,B). Finally, we investigated the ability of neutrophil supernatants to promote nucleic acid sensing. Surprisingly, supernatants from unstimulated neutrophils incubated with poly(I:C) promoted Mx1 expression, whereas PMA-activated neutrophil supernatants elicited poor Mx1 expression in response to poly(I:C). In the absence of poly(I:C), no induction could be reported (Figure 7C).