Source and culture of mycobacteria

All used mycobacterial strains are shown in Table 1. Mycobacteria were amplified, stored and cultured as described previously [33]. Representative vials were thawed and enumerated for viable CFU on Middlebrook 7H11 plates (BD Biosciences). Live-dead staining (BacLight, Invitrogen) of bacterial suspensions with fluorochromic substrates revealed a viability of the bacteria > 90%. rior to use, aliquots were sonicated in a water bath for 10 min at 40 kHz and 110 W at room temperature to disrupt small aggregates of bacteria. The cell culture of mycobacteria occurred on Middlebrook 7H11 agar plates. 2 × 106 sonicated mycobacteria were disseminated on the plates in variant logarithmic dilutions (1:10, 1:100 and 1:1000). Then an incubation occurred for 14 days at 37 °C. After the incubation, the colonies were CFU.

Table 1 Mycobacterial strains 3 H-Uracil proliferation assay

As a correlate of mycobacterial viability, we measured the incorporation of radioactively-labelled 5.6-3H-uracil (ART-0282, Biotrend, Cologne, Germany) into the bacterial RNA [33]. 2 × 106 sonicated mycobacteria were incubated with the peptides in Middlebrook 7H9 broth in a 96-well plate. All samples were set up in triplicate, using 2 µg/ml rifampin (Mtb) or 2 µg/ml clarithromycin (NTM) as controls. 3H-Uracil (0.3 µCi/well) was added after 72 h for Mtb or 48 h for NTM and cultures were incubated for an additional 18 h. Mycobacteria were then inactivated by treatment with 4% paraformaldehyde (PFA, Sigma-Aldrich) for 30 min and transferred onto glass fiber filters (Printed Filtermat A, PerkinElmer) using a 96-well based Filtermat Harvester (Inotech). Fiber filters were dried in a microwave at 240 W for 5 min and sealed at 75 °C with a sheet of solid scintillant wax (MeltiLex, PerkinElmer). Radioactivity was measured using a β-Counter (Sense Beta, Hidex). Antimicrobial activity (%) was calculated as counts per minute (cpm) of the treated sample/cpm of the un-treated sample × 100.

Infection of macrophages with Mtb

As described previously, human peripheral blood mononuclear cells (PBMC) were isolated from buffy coats of anonymous donors (Institute of Transfusion Medicine, Ulm University) by density gradient centrifugation (Ficoll-Paque Plus, GE Healthcare). Monocytes were selected from plastic adherence and thoroughly washed. For generation of monocyte-derived macrophages, cells were cultured in M-SFM with granulocyte–macrophage colony-stimulating factor (GM-CSF, 10 ng/ml, Miltenyi) for 6 days as described previously [33]. Afterwards, macrophages were detached using EDTA (1mM, Sigma-Aldrich) and bulk-infected in 6-well plates with single-cell suspensions of Mtb at a multiplicity of infection of 5. After 2 h, cells were thoroughly washed to remove extracellular bacteria and harvested using 1 mM EDTA (Sigma-Aldrich). The infection rate was regularly determined by auramine-rhodamine staining and showed a donor-dependent variability between 17 and 32%. The number of bacilli per infected macrophages was within the range of 1–3.

Quantification of intracellular mycobacterial growth: mycobacterial growth inhibition assay (MGIA)

Macrophages were bulk-infected with Mtb (MOI = 5, 2 h) and extracellular bacilli were removed by thorough rinsing, harvested and seeded in a 24-well plate at a concentration of 1 × 105 cells in 300 µl M-SFM. Infected macrophages were treated as indicated and after 72 h cells were lysed by adding 200 µl of water. The lysates were transferred to mycobacteria growth indicator tubes (MGIT, BD Biosciences, Franklin Lakes, USA), supplemented with 800 µl supplement containing oleic acid, albumin, dextrose and catalase (OADC, BD Biosciences) and incubated in a Bactec MGIT 320 system (BD Biosciences). The Bactec MGIT 320 detects oxygen consumption and gives the exact time (in minutes) from the beginning of culture to the detection of bacterial activity (time to positivity). The number of viable bacilli was calculated by comparison of the time to positivity in the sample with a standard curve obtained from the growth of tenfold diluted mycobacterial suspensions (103 to 107 bacilli, duplicate).

Peptide origin and synthesis

Lysozyme was extracted and purified from human neutrophil granulocytes (Sigma-Aldrich). Lys-H1 (KVFERCELARTLKRLGM) was synthesized by PSL Heidelberg (Heidelberg, Germany) using F-moc chemistry [33]. For visualization of Lys-H1, the peptide was conjugated N-terminally to the fluorescent dye Atto647N (PSL Heidelberg). Peptides were purified to > 95% homogeneity by reverse-phase HPLC. To allow a valid comparison between the experiments, all concentrations are given in molarity. This partially resulted in uneven concentrations and non-linear titration steps.

Detection of endogenous lysozyme in macrophages

Mtb-infected macrophages were distributed on 8-chamber slides (Thermo Fisher Scientific; final volume 100 µl). After fixation of the cells with PFA (Sigma; final concentration 4%), non-specific binding sites are blocked using a blocking buffer (1 h at room temperature). This additionally permeabilizes the cell wall for intracellular staining. The antibodies are diluted and pipetted one after the other in a dark chamber (1 h incubation each). After staining, the slides were examined under the microscope for lysozyme, Mtb, and a colocalization of both.

All used antibodies are shown in Table 2. Quantification of colocalization was determined by scanning 10 fields of view per experiment with the software ZEN (Zeiss) and ImageJ (Wayne Rasband).

Uptake of Lys-H1 in macrophages

To assess the uptake of Lys-H1 in macrophages, the peptide was labeled with the fluorescent Atto647N by PSL Heidelberg. 1 × 105 macrophages were incubated with 15 µM Lys-H1-Atto in an 8-chamber-slide, unstimulated macrophages were used as negative control. After 24 h, the supernatant was removed and cells were fixed using 4% paraformaldehyde (PFA), followed by blocking in 0,5% bovine serum albumin (BSA). Cell nuclei were stained with DAPI (5µg/ml). Representative pictures of five different donors were acquired using the inverted laser scanning confocal microscope LSM 710 (Zeiss, Oberkochen, Germany). Image processing was performed using ImageJ software (v 1.52c). All images displayed in this study were processed for brightness/contrast. Mtb were resuspended in 500 µl dye buffer and washed twice at 10.000 rpm for 10 min. The pellet was resuspended in 50 µl dye buffer and 5 µl Succinimidyl-Ester-488 (invitrogen, Waltham, USA) was added for one hour. In a 24-well plate, about 5 × 105 macrophages were incubated with the succinimidyl-ester stained Mtb for 2 h, using a multiplicity of infection of 12,5. Afterwards, 50 nM Lys-H1-Atto (lower concentration required due to higher sensitivity compared to microscopy) were added for 24 h. The supernatant was transferred into round bottom tubes and the remaining macrophages were removed from the wells using 1mM PBS-EDTA (Thermo Fisher Scientific) and transferred into the corresponding tubes. The uptake of Mtb and Lys-H1-Atto in macrophages was measured using the flow cytometer FACS Canto II (Becton Dickinson, Franklin Lakes, USA) with FlowJo. For investigation of Lys-H1 cell-specific uptake, macrophages were seeded in sterile FACS tubes alongside autologous PBMC in a 1:1 ratio for a total cell count of 0.5 × 106 cells in AIM V medium [33]. Cells were incubated with Lys-H1-Atto647N for 2 h at 37 °C. Afterwards, cells were washed with FACS buffer and centrifuged for 10 min at 1300 rpm. Supernatant was discarded and cells were stained against MHC class II by a FITC-conjugated HLA-DR antibody (1:200). Sample analysis was performed using a FACSCalibur flow cytometer (BD Biosciences). Data analysis was performed using FlowJo Version 10.5.3 (BD Biosciences) and GraphPad Prism Version 8.2.1 (GraphPad Software).

Screening of haemofiltrates

Haemofiltrates of unknown concentration were dissolved in aqua. Starting from thousands of litres of haemofiltrate, a waste product of haemodialysis of patients with renal failure, a peptide library was generated [43]. Peptides were separated into 300–500 pools based on charge (cation-exchange separation) and hydrophobicity (reversed-phase liquid chromatography). This haemofiltrate peptide library is a salt-free source of highly concentrated peptides and small proteins (< 30 kDa) which can be exploited for the unbiased search for antimicrobial peptides (Bosso 2018). Subsequently, an 3H-Uracil proliferation assay was performed with the addition of different haemofiltrates fractions as described above. Two additional rounds of chromatographic purifications were performed to obtain a pure active compound. The active chromatographic fraction from the peptide bank was subjected to reversed-phase chromatography on Source RPC 15 (Cytiva, USA) of dimensions 2 × 25 cm. at a flow rate of 13 ml/min, using the gradient of acetonitrile 0/5, 10/25, 15/30/ 55/50, 65/60 and 75/100 (tun time, min/%B) The second purification step was performed on a Phenomenex Aqua RP18 (Phenomenex, USA) of dimensions 10 × 25 cm at a flow rate of 2.5 mL/min, using the gradient of acetonitrile 0/5, 10/30, 15/35, 55/45, 65/60 and 75/100 (run time, min/%B). A Varian Star chromatographic system (50 mL pump head) was employed for purifications, using an elution system composed of A, 0.1% TFA in water, and B, 0.05%TFA in a mixture acetonitrile/water (4:1, v/v). Detection of eluting compounds was online monitored at 280 nm.

Mass spectrometry analysis

Molecular mass measurement by MALDI-TOF: The sample was analysed by an Axima Confidence MALDI-TOF MS (Shimadzu, Japan) in positive linear mode on a 384-spot stainless-steel sample plate. Spots were coated with 1 µL 5 mg/mL CHCA previously dissolved in matrix diluent (Shimadzu, Japan), and the solvent was allowed to air dry. Then 0.5 µL sample or standard was applied onto the dry pre-coated well and immediately mixed with 0.5 µL matrix; the solvent was allowed to air dry. All spectra were acquired in the positive ion linear mode using a 337-nm N2 laser. Ions were accelerated from the source at 20 kV. A hundred profiles were acquired per sample, and 20 shots were accumulated per profile. The equipment was calibrated with a standard mixture provided in the TOFMixTM MALDI kit (Shimadzu, Japan). Measurements and MS data processing were controlled by the MALDI-MS Application Shimadzu Biotech Launchpad 2.9.8.1 (Shimadzu, Japan).

Sequencing by LC-MS/MS: The sample was reduced with 5 mM DTT for 20 min at RT, then carbamidomethylated with 50 mM iodoacetamide for 20 min at 37 °C, and subsequently digested with trypsin (ThermoFisher Scientific, 900,589), at a 1:50 ratio (enzyme: protein) for 16 h at 37 °C. A 15 µl aliquot of the digested sample was analyzed using an Orbitrap Elite Hybrid mass spectrometry system (Thermo Fisher Scientific) online coupled to an U3000 RSLCnano (Thermo Fisher Scientific) uPLC as described [40]. XCalibur 2.2 SP1.48 (Thermo Fisher Scientific, Bremen, Germany) was used for data-dependent tandem mass spectrometry (MS/MS) analyses. Database search (PEAKS DB) was performed using PEAKs X + studio [55]. For peptide identification, MS/MS spectra were correlated with the UniProt human reference proteome set (Uniprot release 2023_03; 20,423 reviewed entries). Parent mass error tolerance and fragment mass error tolerance were set at 15 ppm and 0.5 Da, respectively. Maximal number of missed cleavages was set at 3. Carbamidomethylated cysteine was considered as a fixed modification, and methionine oxidation as a variable modification. False discovery rates were set on the peptide level to 1%.

Statistical analysis

All statistical analyses as mentioned in the figure legends were performed using GraphPad Prism v8.2.1 (GraphPad Software). Significance was calculated using non-parametric tests for paired samples (Wilcoxon-Rank test, paired t-test). Differences were considered significant when p-value < 0.05.