Activity-based profiling of neutrophil serine proteases: detection, epithelial interactions, and antiviral potential Akmaral Assylbekova A thesis submitted in partial fulfillment of the requirement of Nazarbayev University for the degree of Doctor of Philosophy in Life Sciences May 2025 Supervisory Committee Lead supervisor: Professor Timo Burster (Nazarbayev University) Co-supervisor 1: Associate Professor Ferdinand Molnar (Nazarbayev University) Co-supervisor 2: Associate Professor Jeanette Kunz (Nazarbayev University) External co-supervisor: Professor Reinhold Schirmbeck (Ulm University) 2 Examination Committee The Ph.D. thesis of Akmaral Assylbekova has been approved by the examination committee. Committee Chair: Professor Timo Burster (Nazarbayev University) Committee members: Associate Professor Davide Antonio Abate (University of Padua), Associate Professor Luca Vangelista (University of Pavia), Associate Professor Tursonjan Tokay (Nazarbayev University) 3 Copyright © 2025 Akmaral Assylbekova All rights reserved 4 Abstract Activity-based profiling of neutrophil serine proteases: detection, epithelial interactions, and antiviral potential Akmaral Assylbekova Neutrophils, which are immune cells involved in innate immunity, secrete neutrophil serine proteases (NSPs) into the extracellular environment during inflammation. Cathepsin G (CatG), neutrophil elastase (NE), proteinase 3 (PR3), and neutrophil serine protease 4 (NSP4) represent NSPs, which are crucial components of immune defense and the inflammatory response. NSPs not only act as antibacterial agents but also play a critical role in antiviral immunity. Serine proteases, including NSPs, are abundant in the respiratory tract during inflammation and are responsible for degrading viral-derived glycoproteins, thereby preventing potential infection. Despite extensive research on the enzymatic functionality of NSPs, the interaction with epithelial surfaces for possible viral defense remains inadequately characterized. In the following thesis, the binding capacity of NSPs, including CatG and NE, to lung epithelial cells and their proteolytic activity were investigated. Employing activity- based protein profiling (ABPP) with novel fluorophore-conjugated activity-based probes (ABPs) to detect the catalytic activity of NSPs in SDS-PAGE-, flow cytometry-, and confocal microscopy assays. The specificity of ABPs was verified by applying different selective serine protease inhibitors. Furthermore, the serine protease inhibitor camostat, currently undergoing clinical trials to mitigate the effects of coronavirus disease 2019 (COVID-19), did not inhibit CatG, NE, or PR3. Moreover, catalytically active NSPs can bind to the cell surface of lung epithelial cell lines and were internalized into the cells, which was tracked by using ABPs. This suggests that epithelial-bound NSPs may serve as pivotal immune guardians and enhance the efficacy of neutrophil-driven pathogen eradication by proteolytically degrading invading microbes at the epithelial cell surface, thereby potentially fortifying host defense at mucosal surfaces. Further research is needed to understand the physiological- but also pathological consequences of epithelial-bound NSPs; particularly concerning respiratory infections caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). In conclusion, NSPs emerge as a promising target for therapeutic interventions, especially through the development of small molecule modulators that enhance the catalytic activity of NSPs to improve viral clearance. 5 List of Tables Table 1. Components for 12% and 5% SDS polyacrylamide gels. Table 2. Cell lines. Table 3. Software and online tools. Table 4. Molecular docking results of camostat docked to TMPRSS2 (PDB ID 7MEQ), CatG (PDB ID 1CGH, NE (PDB ID 1B0F), and PR3 (PDB ID 1FUJ). List of Figures Fig 1. The Schechter–Berger (1967) nomenclature for substrate recognition by proteases. Fig 2. A schematic overview of a peptidyl diphenyl phosphonate activity-based probe (ABP), illustrating its chemical structure, mechanism of action, and common analytical applications in activity-based protein profiling (ABPP). Fig 3. Flow cytometry gating strategy for the detection of surface-bound CatG. Fig 4. Activity-based protein profiling (ABPP) of CatG, NE, and PR3 using SDS-PAGE and in-gel fluorescence detection system. Fig 5. Comparative titration of MARS116-FAM and MARS116-FAM ∆C5. Fig 6. Sivelestat inhibits NE and PR3. Fig 7. Camostat does not inhibit NSP activity. Fig 8. CatG enzymatic activity assay using colorimetric substrate. Fig 9. Binding poses of CatGinh and camostat predicted by AutoDock 4. Fig 10. Binding poses of sivelestat and camostat docked into NE and PR3 as predicted by AutoDock4. Fig 11. Predicted binding pose of camostat in the active site of TMPRSS2 using AutoDock4. 6 Fig 12. Detection of CatG on the surface of lung epithelial cells using flow cytometry. Fig 13. Effects of serum containing medium to localize proteolytically active CatG. Fig 14. Fluorescence intensity of MARS116-FAM-treated cells under serum-free and FBS- supplemented culture and inhibition conditions. Fig 15. Time-course incubation of CatG-ABP complex in A549 and H1299 cells. Fig 16. Specificity of CatG labeling in H1299 cells. Fig 17. Visualization of NE in A549 cells. Fig 18. Dual roles of neutrophil serine proteases (NSPs) in SARS-CoV-2 infection and inflammation. List of Abbreviations α1AT α1-antitrypsin ABP Activity-based probe ABPP Activity-based protein profiling ACE2 Angiotensin-converting enzyme 2 ACT α1-antichymotrypsin APC Antigen-presenting cell ARDS Acute respiratory distress syndrome ATII Alveolar type II BALF Bronchoalveolar lavage fluid CatG Cathepsin G CatGinh Cathepsin G inhibitor I 7 CF Cystic fibrosis COPD Chronic obstructive pulmonary disease COVID-19 Coronavirus disease 2019 CTCF Corrected total cell fluorescence CyTOF Mass cytometry by time-of-flight DAP α-Aminoalkylphosphonate diphenyl esters DAP22c Bt-Suc-Ala-Ala-PheP(OPh)₂ (Biotinylated diphenyl phosphonate probe) DAPe 1.5-diaminopentane DPPI Dipeptidyl peptidase I (cathepsin C) ELISA Enzyme-linked immunosorbent assay ECM Extracellular matrix ESI Electrospray ionization FAM 5(6)-Carboxyfluorescein FITC Fluorescein isothiocyanate FRET Förster Resonance Energy Transfer HIV-1 human immunodeficiency virus type 1 IBD Inflammatory bowel disease IL-7 Interleukin 7 IL-8 Interleukin 8 LF Lactoferrin LTB4 Leukotriene B4 8 mAbs monoclonal antibodies MALDI Matrix-assisted laser desorption/ionization MARS116-FAM MARcin Sienszyk FAM-DAPe-Suc-Val-Pro-PheP(OPh)2 (Cathepsin G-specific fluorescent ABP) MARS116-FAM* FAM-DAPe-Suc-Val-Pro-Phe-COO-, inactive control version of MARS116-FAM (non-reactive warhead) MARS116-FAM ∆C5 Short-linker variant of MARS116-FAM MHC I Major histocompatibility complex class I MS Mass spectrometry NE Neutrophil elastase NETs Neutrophil extracellular traps NK Natural killer cells NSPs Neutrophil serine proteases NSP4 Neutrophil serine protease 4 PPI Protein-protein interaction PR3 Proteinase 3 qABP Quenched activity-based probe RSV Respiratory syncytial virus RT Room temperature SARS-CoV-2 Severe acute respiratory syndrome coronavirus 2 SCCA2 Squamous cell carcinoma antigen 2 SD Standard deviation 9 SDS-PAGE Sodium dodecyl sulfate–polyacrylamide gel electrophoresis SEM Standard error of the mean Serpins SERine Protease INhibitors SFTI Sunflower trypsin inhibitor SLPI Secretory leukocyte protease inhibitor STED Stimulated emission depletion STR Short tandem repeat SucVPF Suc-Val-Pro-Phe(OPh)₂ TMPRSS2 Transmembrane Protease, Serine 2 VPV-FAM FAM-DAPe-Suc-Val-Pro-ValP(OPh)2, NE/PR3-targeting fluorescent ABP Y2H Yeast two-hybrid system 10 Acknowledgments I would like to begin by acknowledging my supervisory and examination committees for their guidance, support, and evaluation of my work. My deepest gratitude goes to my lead supervisor, Professor Timo Burster. From the very beginning, he made me feel that I belong in research. He teaches with heart and shares knowledge often without even realizing it. He welcomed me as his student despite being cautioned that I was a dedicated mother, and that gesture meant a great deal to me. During the COVID-19 lockdown, he ensured that I could continue my work, which led to our first paper, where I had the privilege of being first author. I am equally grateful to Ferdinand Molnár for his constant support, whether through sharing knowledge, providing computing resources, offering his positivity, or giving a thoughtful review of my thesis that greatly strengthened the final work. I feel fortunate to have been mentored by both of them. I remain especially grateful to Dr. Christian Schoenbach, who witnessed the beginning of this project but sadly did not live to see its completion. His dedication to the program and his support of me as a student are fondly remembered. I am deeply thankful to my lab group and colleagues, especially the labs of Prof. Dos Sarbassov and Prof. Tri Pham, for their constant backup, guidance, and support in countless daily routines. It was a privilege to work alongside them, and I am especially grateful for their help in running experiments, sharing knowledge, and ensuring that this research had the resources it required. A special mention also goes to the BioCluster WhatsApp group, whose quick advice, humor, and encouragement often provided the perfect mix of science and support at just the right moments. I also thank the Nazarbayev University Core Facilities for their support with confocal microscopy and flow cytometry. I would also like to acknowledge my friends, who made sure I was well-rested and kept life balanced with many activities outside of work. Their presence reminded me that joy, laughter, and breaks were just as important as long hours in the lab. I am especially indebted to my family. My deepest gratitude is to my Mother - Aliya, who supported me tirelessly throughout these years. She ensured that my children were cared for and helped me immensely while I worked long hours, both day and night. Without her, this journey would not have been possible. My sister, Alina, paved the path of PhDs in our family and offered guidance and encouragement that inspired me to persevere. My husband, Zhenisbek, provided unwavering support, understanding, and positivity, always lifting my spirits when this journey felt overwhelming. And of course, my daughters, Inzhu and Deniza, who have been with me throughout this path—Inzhu since my Master’s years and Deniza, who was born during the first year of my PhD. They made me stronger, more resilient, and ultimately shaped the person I am today. Finally, I want to express my heartfelt gratitude to my aunt Zoya. From my childhood, she showed me how important research is and encouraged me to pursue this path. Her early influence played a meaningful role in guiding me toward this career. 11 Declaration I declare that the research contained in this thesis, unless otherwise formally indicated within the text, is the original work of the author. The thesis has not been previously submitted to this or any other university for a degree and does not incorporate any material already submitted for a degree. I confirm that content previously published in open access journals has been reused in accordance with the journal’s licensing terms, which permit reproduction with appropriate citation and attribution. Parts of this thesis include figures and text adapted from previously published work (Assylbekova et al., 2021 and 2024). These articles were published under open access Creative Commons CC BY license, therefore no additional permission from the journal was required. Signature Date 12 Table of Contents Examination Committee ........................................................................................................ 2 Copyright © 2025 .................................................................................................................. 3 Abstract ................................................................................................................................... 4 List of Tables .......................................................................................................................... 5 List of Figures ......................................................................................................................... 5 List of Abbreviations ............................................................................................................. 6 Acknowledgments ................................................................................................................ 10 Declaration ............................................................................................................................ 11 Chapter 1 Literature review ............................................................................................... 15 1.1 The evolution of protein science to functional proteomics ..................................... 15 1.1.1 Early studies on proteins ................................................................................... 15 1.1.2 Transition from genomics to proteomics .......................................................... 15 1.1.3 Technological advances in proteomics ............................................................. 15 1.1.4 Emergence of chemical proteomics .................................................................. 16 1.2 Introduction to proteases, serine proteases, and neutrophil serine proteases ...... 16 1.2.1 Overview of proteases and structure/catalysis of serine proteases ................... 16 1.2.2 The Roles of neutrophil serine protease in the immune system ....................... 18 1.2.3 The role of NSPs in viral defense ..................................................................... 20 1.2.4 Detection and quantification of NSP activity ................................................... 21 1.3 Activity-based protein profiling (ABPP) for protease detection ........................... 22 1.3.1 Introduction to ABPP ........................................................................................ 22 1.3.2 Core methodologies of ABPP ........................................................................... 22 1.3.2.1 Activity-dependent labeling of proteases ................................................... 22 1.4 Current ABPs for the detection of neutrophil serine proteases ............................ 26 1.4.2 Relevance of ABPs to neutrophil function and disease states .......................... 27 1.4.3 Development of ABPs for NSPs ...................................................................... 27 1.4.3.1 Development and structure of ABPs to detect active CatG ....................... 28 1.4.3.2 Other NSP ABPs ........................................................................................ 29 1.4.4 Application of MARS116 in flow cytometry assays ........................................ 29 1.4.5 Advantages and future perspectives of MARS116-FAM ................................. 30 1.5 Study rationale, aims, objectives, and hypotheses .................................................. 32 Hypothesis: ABPs can specifically and sensitively detect active proteases ................. 32 13 Hypothesis: Neutrophil proteases CatG and NE might bind to lung epithelial cell surfaces in an active manner ........................................................................................... 33 Hypothesis: Cellular conditions, such as nutrient availability (e.g., serum starvation), could influence protease binding/activity on epithelial surfaces. ............ 34 Hypothesis: MARS116-FAM ∆C5 might show superior sensitivity for membrane bound CatG compared to MARS116-FAM. .................................................................. 34 Hypothesis: VPV-FAM might label NE and PR3 with shared specificity .................. 35 Hypothesis: Camostat mesylate might inhibit NSPs in a dose-dependent manner. .. 35 Chapter 2 Materials and Methods ...................................................................................... 37 2.1 Materials ..................................................................................................................... 37 2.1.1 Chemicals and consumables ............................................................................. 37 2.1.2 Activity-based probes ....................................................................................... 37 2.1.3 Chemical inhibitors ........................................................................................... 38 2.1.4 Laboratory equipment ....................................................................................... 38 2.2 Methods ....................................................................................................................... 39 2.2.1 Sodium dodecyl sulfate - polyacrylamide gel electrophoresis (SDS-PAGE) coupled with activity-based protein profiling (ABPP) .............................................. 39 2.2.2 Cell culture and maintenance ............................................................................ 41 2.2.3 Measurement of membrane-associated CatG activity by flow cytometry ........ 43 Sample preparation ................................................................................................ 43 Labeling of CatG with probes ................................................................................ 44 Data acquisition and analysis ................................................................................. 44 2.2.4 Confocal microcopy .......................................................................................... 45 Probe titration and protease labeling ...................................................................... 47 Inhibitor titration assay .......................................................................................... 47 2.3 Molecular docking methodology .............................................................................. 48 2.3.1 Molecular docking parameters .......................................................................... 48 2.3.2 Validation of the docking protocol ................................................................... 49 2.4 Statistical methods ..................................................................................................... 49 2.5 Software and online tools .......................................................................................... 50 Chapter 3 Results ................................................................................................................ 52 3.1 Validation and optimization of activity-based probes ............................................ 52 14 3.1.1 Gel-based ABPP confirms probe sensitivity ..................................................... 52 3.1.2 Comparative analysis of MARS116-FAM and MARS116-FAM ∆C5 ............ 53 3.2 Competitive inhibition assays ................................................................................... 54 3.2.2 Sivelestat inhibits NE and PR3 ......................................................................... 55 3.3 Effects of camostat on CatG, NE, and PR3: “Camostat does not inhibit NSP activity” (Assylbekova et al., 2021) ................................................................................. 56 3.4 Molecular docking ..................................................................................................... 58 3.4.1 The camostat binding pose in CatG vs. known inhibitors of CatG ................. 59 3.4.2 Camostat binds to the peripheral region of the catalytic center of NE ............. 61 3.4.3 Camostat has a weak engagement with PR3 in contrast to sivelestat ............... 62 3.5 Cell-based ABPP ........................................................................................................ 63 3.5.1 Flow cytometry ................................................................................................. 63 3.5.2 Confocal microscopy ........................................................................................ 65 3.6 Summary of key findings in the results section ....................................................... 73 Chapter 4 Discussion. .......................................................................................................... 74 4.1 NSPs in immune defense with antiviral and proinflammatory roles .................... 74 4.2 The properties of ABPP, inhibitors, and camostat ................................................. 80 4.3 Conclusions ................................................................................................................. 83 4.4 Future directions and therapeutic applications ...................................................... 84 References ............................................................................................................................. 86 Journal publications ............................................................................................................ 97 Appendix A. Short tandem repeat profiling ...................................................................... 98 Appendix B. Supplementary methods ............................................................................. 100 Appendix C. Extended data and Figures ......................................................................... 102 Chapter 1 Literature review 1.1 The evolution of protein science to functional proteomics 1.1.1 Early studies on proteins The study of proteins began in the 18th century with early observations of their physiochemical properties (e.g., coagulation). The first empirical formula for proteins was proposed by Mulder (1838), who also introduced the concept of amino acids. The term "protein" was first mentioned in Berzelius’ 1838 letter to Mulder (Hartley, 1951). However, their biological function remained unknown until Sumner (1926) demonstrated that enzymes are proteins. The field evolved significantly with breakthroughs, such as sequencing of insulin (Sanger et al., 1953) and the 3D structural characterization of myoglobin and hemoglobin by John Kendrew and Max Perutz (Kendrew et al., 1958; Perutz et al., 1960). 1.1.2 Transition from genomics to proteomics If previous studies concentrated on examining individual proteins, the late 20th century marked the beginning of systematic investigations of cytoskeletal proteins, enzyme cascades, and signaling pathways (Perret, 2007). Functional genomics initially analyzed mRNA expression, but a 1997 study revealed that mRNA and protein levels have a poor correlation (Anderson et al., 1997; Gygi et al., 1999). This finding led to the rise of proteomics (PROTEin+genOME), first termed by Marc Wilkins in 1994 (Wilkins et al., 1996). Unlike genomics, proteomics studies alternative splicing and post-translational modifications (PTMs) that are critical for understanding biological systems (Böttcher and Sieber, 2010). 1.1.3 Technological advances in proteomics In the early days of proteomics, 2D gel electrophoresis was used for protein visualization. Mass spectrometry (MS), especially soft ionization techniques, including electrospray ionization (ESI) and matrix-assisted laser desorption/ionization (MALDI), along with hybrid analyzers, addressed the limitations in resolution (Hu et al., 2005; Han et al., 2008). Further advances in post-translational modifications (PTM) analysis were made with tandem MS (MS/MS) fragmentation techniques (Syka et al., 2004). Functional proteomics expanded with protein-protein interaction (PPI) studies using techniques like protein and antibody microarrays (enzyme-linked immune-sorbent assay, ELISA-based), yeast two-hybrid (Y2H), and phage display (Smith, 1985; Uttamchandani et al., 2006; Fields et al., 1989). Additional 16 approaches that emerged were fluorescent protein fusion, knockout studies, and RNA interference (Van Roessel et al., 2002; Zambrowicz et al., 1998; Pelletier et al., 2010). 1.1.4 Emergence of chemical proteomics To study protein function beyond protein expression, chemical proteomics integrated chemistry and biology, employing chemical tagging, such as isotope labeling and isobaric tagging (Ong et al., 2002; Ross et al., 2004) for quantitative analysis. PTM enrichment strategies focus on phosphorylation and glycosylation (Sun et al., 2005). Other chemical proteomics applications include small-molecule interaction studies and drug discovery via chemical microarrays (Schirle et al 2012; Ma et al., 2006). As proteomics evolved from structural characterization to functional analysis, it became evident that understanding protein activity and function, rather than just abundance, was critical to understand biological mechanisms. Among the vast repertoire of functional proteins, enzymes, particularly proteases, emerged as key regulators of cellular processes. Studying proteases and their precise enzymatic activity became a focal point in functional proteomics. Eventually, this necessitated the development of specialized techniques such as activity-based protein profiling (ABPP) to directly assess the catalytic function of proteases in biological systems (Sanman and Bogyo, 2014). 1.2 Introduction to proteases, serine proteases, and neutrophil serine proteases 1.2.1 Overview of proteases and structure/catalysis of serine proteases Proteases, also known as peptidases or proteinases, are indispensable for numerous biological processes, including proteolytic degradation of proteins, signal transduction, immune regulation, and tissue remodeling (Turk et al., 2012). In humans, the genome encodes around 550 proteases, representing nearly 2% of all genes, underscoring their biological significance (López-Otín and Bond, 2008; Edgington et al, 2011). Proteases are classified based on their catalytic mechanisms, which determine how they hydrolyze peptide bonds. Among the five major classes, metallo-, aspartic-, cysteine-, threonine-, and serine proteases, each has a distinct catalytic mechanism (Hedstrom, 2002). Among these, serine proteases stand out due to their highly efficient and specific catalytic activity, facilitated by a conserved catalytic triad composed of serine (Ser), histidine (His), and aspartate (Asp) (Barret and Rawlings, 2001; Hedstrom, 2002). Representing about one- third of all proteases, serine proteases regulate diverse physiological processes, such as 17 digestion (trypsin, chymotrypsin), coagulation (thrombin, factors VIIa, IXa, and Xa), fibrinolysis (plasmin, urokinase), and immune defense (neutrophil serine proteases) (Grzywa and Sienczyk, 2013). The variety in structure and function positions the enzyme class as one of the most widely researched, having significant consequences for both health and disease. Under physiological conditions, serine proteases operate through a two-step catalytic mechanism: 1. Nucleophilic attack and acyl-enzyme formation: the hydroxyl group of the catalytic Ser acts as a nucleophile, attacking the carbonyl carbon of the peptide bond. This results in a tetrahedral intermediate which collapses to release the first peptide fragment and forms a covalent acyl-enzyme complex. 2. Deacylation and product release: a water molecule, activated by catalytic triad His, hydrolyzes the acyl-enzyme intermediate, leading to the release of the second peptide fragment and regeneration of the active enzyme for subsequent catalysis (Hedstrom, 2002). Before discussing the specificity of serine proteases, it is essential to define how proteases recognize and cleave the substrate. The specificity of serine proteases is determined by the structure of the active site, particularly the substrate-binding subsite. This is best described using the Schechter and Berger (1967) nomenclature, a widely adopted system defining subsite interactions in protease-substrate recognition (Fig 1). The peptide sequence is designated as …-P3-P2-P1-P1’-P2’-P3’-…, corresponding directly to the subsites within the protease active site, which are labeled as …-S3-S2-S1-S1’-S2’-S3’-… The cleavage occurs between the P1 and P1’ positions, at the scissile bond. The P1 residue is the key determinant of protease specificity, as it directly interacts with the S1 pocket of the enzyme. Therefore, P1 dictates the substrate preference of the protease. 18 Fig 1. The Schechter–Berger (1967) nomenclature for substrate recognition by proteases. A diagram illustrating the interaction between the amino acids of the substrate adjacent to the cleavage site (scissile peptide bond) (P) and the corresponding protease subsites near the active site (S). Modified from Wysocka and Lesner (2013). Serine proteases exhibit a remarkable specificity, directed by the structure of their active site and substrate-binding pockets. For example: ● Trypsin prefers positively charged residues like lysine (Lys) and arginine (Arg) at P1. ● Chymotrypsin selects for aromatic residues such as phenylalanine (Phe), tyrosine (Try), and tryptophan (Trp). ● Elastase favors small, non-polar residues like alanine (Ala) and valine (Val). 1.2.2 The Roles of neutrophil serine protease in the immune system Neutrophils, the most abundant human white blood cells, are equipped with a specialized arsenal of neutrophil serine proteases (NSPs), stored in azurophilic granules, and released into the infected or inflamed tissues upon neutrophil activation. These proteases play key roles in innate immunity, inflammation, and host defense by direct pathogen clearance, modulating immune response, extracellular trap formation (NETosis), and facilitating tissue remodeling (Fourschou and Borregard, 2003; Korkmaz, 2008) The four NSPs identified in humans include: cathepsin G (CatG), neutrophil elastase (NE), proteinase 3 (PR3), and neutrophil serine protease 4 (NSP4) (Burster et al., 2021). Although NSPs share a conserved catalytic mechanism, they exhibit distinct substrate specificities and biological functions (Kasperkiewicz, 2021). CatG prefers bulky/aromatic residues (Phe, Tyr, and Lys or Arg with lower efficacy) and Leu in the P1 position, Pro at P2; NE and PR3 prefer sequences with small aliphatic residues (Ala, Val) in P1 position (Wysocka et al., 2007; Bode et al., 1989; Rao et al., 1991). NSPs are synthesized as zymogens to prevent premature activation and undergo maturation via proteolytic cleavage by dipeptidyl peptidase I (DPPI, cathepsin C) in neutrophil granules (Neurath, 1984; Adkison et al., 2002). 19 Additional control mechanisms by which proteases are tightly regulated include endogenous inhibitors, such as serpins (SERine Protease INhibitors) (e.g., alpha1-antitrypsin; A1AT, secretory leukocyte inhibitor; SLPI, a1-antichymotrypsin; ACT, and squamous cell carcinoma antigen 2; SCCA2), and prevent collateral damage (Ye and Goldsmith, 2001; Korkmaz et al., 2010). Additionally, a context-dependent neutrophil activation is essential for ensuring an effective and balanced immune response. For example, during an infection, neutrophil degranulation leads to rapid release of granule-stored proteases, which help eliminate pathogens while simultaneously regulating inflammation to prevent excessive tissue damage. Once released, NSPs play a dual role in host defense and inflammation. They can directly degrade microbial proteins in a proteolytic manner, thereby killing bacteria and inactivating viruses, and can also modulate immune signaling by processing cytokines/chemokines and cell surface receptors. This proteolytic activity is important for host defense, as CatG and NE support the clearance of bacteria in the airways and can degrade viral coat proteins, contributing to antiviral immunity by inhibiting viral replication (Leborgne et al., 2024). However, NSPs are double-edged swords: while essential for pathogen defense, their uncontrolled proteolytic activity can lead to excessive tissue damage. The lung is particularly vulnerable, as uncontrolled CatG, NE, or PR3 activity can degrade extracellular matrix components in the airways and alveoli, contributing to the progression of inflammatory lung diseases, such as chronic obstructive pulmonary disease (COPD), acute respiratory distress syndrome (ARDS), and emphysema (Burster et al., 2021). NE, in particular, cleaves elastin and a broad range of bacterial proteins, contributing to pathogen elimination within phagolysosomes and extracellularly, yet excess NE activity leads to host tissue damage (Anderson et al., 2019; Korkmaz et al., 2008). Similarly, while PR3 also participates in microbial clearance and processing of cytokines and chemokines, it can also serve as an autoantigen in diseases like Wegener’s granulomatosis (Korkmaz et al., 2008). Similarly, dysregulated NSP activity also contributes to chronic inflammatory and autoimmune conditions, including sepsis and inflammatory bowel disease (IBD) (Korkmaz et al., 2010). In these pathological conditions, neutrophils infiltrate affected tissues, releasing large amounts of NSPs, overwhelming local inhibitors and result in uncontrolled proteolysis and tissue destruction. If endogenous inhibitors, such as α1-antitrypsin (A1AT) and secretory leukocyte protease inhibitor (SLPI) are depleted or ineffective, unopposed NSP activity exacerbates disease progression (Korkmaz et al., 2010). Given their pivotal roles, CatG, NE, 20 and PR3 are considered potential therapeutic targets for inflammatory diseases. This delicate balance highlights the importance of monitoring and understanding NSP activity in physiological and pathological contexts. 1.2.3 The role of NSPs in viral defense NSPs not only function as antibacterial agents but also play an important role in antiviral immunity (Voynow and Shinbashi, 2021; Burster et al., 2021). However, little is known about their antiviral functions with very few studies in the literature. Serine proteases, including NSPs, are abundant in the respiratory tract and often exploited by viruses for glycoprotein processing. Although NSPs can directly process viral proteins, they can indirectly modulate immune responses to enhance antiviral defense (Pham, 2006; Heutinck et al., 2010; Deng et al., 2020; Leborgne et al., 2024; Lopes et al., 2022). Recent studies have highlighted their role in countering respiratory viruses, such as SARS- CoV-2 and respiratory syncytial virus (RSV) (Deng et al., 2020; Leborgne et al., 2024; Lopes et al., 2022). Several studies have demonstrated that NSPs degrade viral surface glycoproteins, which impair viral entry and infectivity. Experimental data suggest that SARS- CoV-2 Spike (S) protein contains multiple cleavage sites for NSPs, making them potent inhibitors of viral fusion by disrupting to bind the angiotensin-converting enzyme 2 (ACE2) receptor (Mustafa et al., 2022; Schilling et al., 2022; Leborgne et al., 2024). CatG and NE have been shown to cleave the S protein of SARS-CoV-2 thereby impairing the binding to the ACE2 receptor (Leborgne et al., 2024). In vivo studies further support this protective role, showing higher viral titers and increased inflammation in neutrophil-deficient mice (Leborgne et al., 2024). Similarly, NE and PR3 target the fusion (F) protein of RSV, preventing viral attachment and reducing infectivity (Lopes et al., 2022). Furthermore, NET-associated NE exhibits strong antiviral properties by clearing RSV (Deng et al., 2020). NE and PR3 integrated in the NETs trap and neutralize viral particles (Voynow and Shibnashi, 2021). However, it is important to note that some viruses, such as human adenoviruses, have evolved mechanisms to evade NSP- mediated degradation, as they exploit NSPs to cleave surface receptors, enhancing epithelial transduction (Readler et al., 2021). In such cases, internalized NSPs may still damage viral proteins after transient membrane binding. Beyond direct viral degradation, NSPs regulate the innate and adaptive immune response by promoting cytokine release and immune cell recruitment. Particularly, NE and PR3 activate 21 interleukin 7 (IL-7), IL-8, and leukotriene B4 (LTB4), which are key neutrophil chemoattractants, amplifying neutrophil recruitment to infection sites (Readler et al., 2021; Cheetham et al., 2024). CatG further modulates immune response by cleaving cytokine precursors. NSPs enhance antigen presentation by increasing the expression of major histocompatibility complex (MHC I) molecules on cell surfaces. MHC I presents viral antigens to CD8+ T cells for immune surveillance, followed by apoptosis to halt productive infection (Assylbekova et al., 2024). However, imbalances in protease-antiprotease interplay and cytokine storms are correlated with elevated NSP levels, even though NSPs contribute to antiviral defense. Excessive NSP activity causes lung injury and exacerbate clinical symptoms in diseases such as COVID-19 (Schimke et al., 2020; McElvaney et al., 2020). 1.2.4 Detection and quantification of NSP activity Conventional ELISA or Western blot assays are commonly used to detect NSPs, particularly NE. Nevertheless, ELISA and Western blot are constrained by its capacity to quantify only total enzyme levels rather than assessing enzymatic activity, thereby rendering such methodologies functionally limited (Liu et al., 2018). To assess NSP activity, FRET-based probes offer higher sensitivity than chromogenic substrates by allowing ratiometric detection (fluorescence ratio of donor and acceptor signals) and eliminating concentration-dependent variability (Gehrig et al., 2012). The latest quantum dot-based FRET probes further improve detection sensitivity, reaching picomolar limits while enhancing photostability and biocompatibility (Rodriges-Rios et al., 2022). Conventional methods, Western blot and ELISA, are not suitable to distinguish between active and inactive NSPs, as they do not account for zymogens or inhibitor-bound proteases, limiting their clinical relevance. This is particularly evident in IBD, where total NE levels fail to correlate with disease severity, whereas activity-based probes (ABPs) selectively detect active NE in inflamed tissue (Anderson et al., 2019). To address these challenges, activity- based protein profiling (ABPP) has emerged as a powerful functional proteomics tool, enabling real-time and selective labeling of catalytically active proteases (Wolf et al., 2015; Burster et al., 2021; Ferguson et al., 2022). ABPP allows precise tracking of enzyme function in health and disease, offering a more accurate and dynamic approach to biomarker discovery and therapeutic development. 22 1.3 Activity-based protein profiling (ABPP) for protease detection 1.3.1 Introduction to ABPP ABPP technique originated in the 1970s with the studies on penicillin binding proteins (Blumberg and Strominger, 1974). However, the modern ABPP methodology emerged later, with pioneering research by Joseph Oleksyszyn, who introduced irreversible inhibitors of serine proteases (Oleksyszyn and Powers, 1989). The contemporary ABPP approach became prominent in the late 1990s through significant contributions by Bogyo et al. (1997) and Liu et al. (1999). Since then, ABPP has been widely adopted in chemical proteomics, enabling direct functional characterization of enzymes. Unlike conventional methods, ABPP selectively labels active enzymes. Additionally, ABPP causes minimal biological disruption, making it a suitable method for live-cell studies, or even whole organism studies without compromising their natural behavior (Blum et al., 2007, Sakamoto & Hamachi, 2023). ABPP enhances drug target identification, pathway analysis, and disease research, and has been successfully applied to report protease activities in diverse biological systems, including cellular models (Garland et al., 2016). 1.3.2 Core methodologies of ABPP The broad applicability of ABPP allows for its integration into various fields, including disease modeling, drug discovery, and functional proteomics. This technique enables a more profound understanding of enzyme regulation under both physiological and pathological conditions. Unlike conventional biochemical assays that measure total protein levels, ABPP uses special probes to selectively label catalytically active enzymes, allowing functional analysis in complex biological environments (Cravatt et al., 2008). Every ABPP experiment comprises two key elements: 1. Activity-dependent labeling, which involves small molecule probes binding selectively to active enzymes. 2. Analytical process (i.e., technique) to detect and characterize the effect of labeling, emphasizing the multidisciplinary nature of ABPP. 1.3.2.1 Activity-dependent labeling of proteases Activity-dependent labeling of proteases employs small synthetic molecule probes designed to selectively bind to the active site, allowing to visualize and quantify active enzymes in 23 complex biological samples (Liu et al., 1999; Cravatt et al., 2008; López-Otín and Bond, 2008). Small synthetic molecule probes covalently bind specifically to the enzyme active sites and are called activity-based probes (ABPs), which distinguish zymogens and inactive protease-inhibitor complexes (Anderson et al., 2019; Sotiropoulou et al., 2022). ABPs are often derived from biologically active natural products or designed as biomimetic analogs of such compounds. These molecules are considered privileged structures due to an evolutionary role in targeting protein functions, which makes selective enzyme inhibition highly effective (Breinbauer et al., 2002). ABPs consist of three essential components (Fig 2): 1. Reactive group (electrophilic warhead): a chemically reactive moiety that forms an irreversible covalent bond with the active enzyme’s nucleophilic residue, such as a serine residue in serine proteases. 2. Linker (spacer): a structural element that separates the reactive group from the reporter tag, optimizing probe solubility, enzyme accessibility, and target selectivity. Usually, it includes a recognition sequence or scaffold that guides the probe to a specific protease (e.g., a peptide or small-molecule motif mimicking enzyme’s natural substrate). 3. Reporter tag: a detectable label used for visualization and quantification of probe- labeled enzymes. These tags commonly include fluorophores (e.g., fluorescein amidite; FAM and rhodamine), affinity tags (e.g., biotin), or radiolabels (Sienczyk and Oleksyszyn, 2009; Burster et al., 2021; Porta and Steel, 2023). 24 Fig 2. A schematic overview of a peptidyl diphenyl phosphonate activity-based probe (ABP), illustrating its chemical structure, mechanism of action, and common analytical applications in activity-based protein profiling (ABPP). (A) The ABP consists of three core components: a reactive diphenyl phosphonate warhead, a tripeptide-based recognition sequence, and a reporter tag (R), such as an affinity tag (biotin) or a fluorophore for detection. The probe irreversibly binds to the serine protease: the active site serine residue performs a nucleophilic attack on the phosphorus atom of the warhead, releasing one phenoxy group (PhOH). A subsequent “ageing” step leads to the loss of the second PhOH group, leaving a covalent adduct. The resulting negatively charged oxygen in the adduct fits into the enzyme’s oxyanion hole, stabilizing the interaction. (B) The resulting covalent enzyme-ABP complex can be analyzed across multiple platforms, including SDS-PAGE with in-gel fluorescence, mass spectrometry, high-dimensional profiling using mass cytometry (CyTOF), flow cytometry, or confocal microscopy. Fluorescent tags allow direct detection of active enzymes by fluorescence gel scanning, flow cytometry, or microscopy, while biotin tag also enables pull-down and identification of targets. Radiolabeled ABPs are being investigated for in vivo imaging, particularly for applications in infection models and the tumor microenvironment (Blum et al., 2007; 25 Aaltonen et al., 2020). These probes enable non-invasive tracking of protease activity, expanding the application beyond in vitro and ex vivo studies. Beyond traditional ABPs, recent advancements include quenched fluorescent ABPs (qABPs), which are non-emissive until they react with the target protease (Blum et al., 2007; Burster et al., 2021). While more common for cysteine protease imaging, adapting qABPs for serine protease imaging is an emerging research focus. A comprehensive review of various types of ABPs is explained in (Fang et al. 2021, Oleksyszyn and Powers, 1991; Blum et al., 2009; Sienczyk and Oleksyszyn, 2009; Chakrabarty et al., 2019; Breidenbach et al., 2020; Kahler et al., 2020). This innovative design and precise labeling mechanism establish ABPs as powerful instruments for investigating enzyme activity within biological systems, understanding disease pathologies, and developing therapeutic and diagnostic intervention strategies. 1.3.2.2 Analytical techniques used in ABPP The analytical techniques used in ABPP vary depending on the experimental goal. Fluorescence imaging enables near-real-time visualization of enzyme activity in live cells and tissues. LC-MS/MS identifies and quantifies active enzymes in complex proteomes, while flow cytometry provides high-throughput detection of enzyme activity at the single-cell level. Western blotting and ELISA-based ABPP allow for quantitative analysis of labeled enzymes, whereas SDS-PAGE is often used to separate and detect probe-labeled proteins. More advanced techniques, such as in vivo imaging using radioisotopes facilitate non-invasive tracking of enzyme activity, and mass cytometry by time-of-flight (CyTOF) enables high- dimensional single-cell enzyme profiling (Burster et al., 2021). Together, the described methods contribute to functional proteomics, disease biomarker discovery, and drug development by providing insights into enzyme function across a range of biological processes. The application of ABPP extends across multiple biomedical research areas. In functional proteomics, ABPP is widely used to define activity profiles in cellular and tissue models, improving the understanding of protein function in different biological states (Blum et al., 2005; Cravatt et al., 2008). In drug discovery, it serves as a key tool in identifying new therapeutic targets, particularly in diseases, such as cancer and inflammatory disorders (Nomura et al., 2010). Additionally, enzyme activity profiling aids in biomarker discovery and diagnostics, where it can detect disease-associated changes in enzyme function, contributing to improved diagnostic accuracy (Edgington et al., 2022; Willems et al., 2014). The use of ABPP for tissue and organelle-specific profiling allows researchers to map enzyme 26 activity within distinct cellular compartments, enhancing the study of localized enzyme regulation (Fang et al., 2021). Beyond mammalian systems, ABPP has been successfully applied in microbiology, where it serves as a powerful tool for studying bacterial pathogenesis and virulence (Krysiak and Sieber, 2016). For a more comprehensive review of the various types of ABPs and applications across different protease classes, readers are referred to excellent reviews available in the literature (Burster et al., 2021; Porta and Steel, 2023; Fang et al., 2021; Kahler et al., 2020). By bridging the gap between genomic data and proteomic function, ABPP provides precise functional characterization of enzymes. Future advancements will likely focus on integrating ABPP with multi-omics approaches for improved biomarker discovery and developing next- generation ABPs with enhanced specificity and imaging capabilities. With these advancements, ABPP is poised to become a cornerstone technique in biomedical research. 1.4 Current ABPs for the detection of neutrophil serine proteases Over the past two decades, a variety of ABPs targeting NSPs have been developed. The most widely used ABPs are peptidyl phosphonate esters, particularly diaryl phosphonates, which form stable covalent adducts with the catalytic serine residue (Oleksyszyn, 1979; Ferguson et al., 2022). These probes have been extensively utilized to study NSP activity, offering insights into neutrophil-mediated inflammation, immune responses, and disease pathology. 1.4.1 Catalytic mechanism and selectivity profile of ABPs for labeling NSPs The effectiveness of an ABP relies on its binding mechanism and the specificity of that interaction. To achieve selectivity, ABPs utilize peptide or non-peptide recognition elements that align with the substrate preferences of the respective proteases. Serine protease-directed ABPs typically employ electrophilic warheads that target the active-site serine hydroxyl group. Among these, diaryl phosphonates are commonly used as they mimic the tetrahedral transition state of peptide hydrolysis, forming a stable probe-enzyme adduct (Grzywa and Sienczyk, 2013). The careful selection of warhead chemistry and recognition sequences ensures that ABPs selectively label active proteases even within the complex proteomes of neutrophils or other cells, tissues, and biological samples. Overall, ABPs bridge a critical gap in neutrophil research, providing molecular detectors to track the “live ammunition” of neutrophils (their active proteases), providing snapshots of activity in a complex environment, thereby enriching our understanding of both normal immune defense and pathological inflammation. 27 1.4.2 Relevance of ABPs to neutrophil function and disease states The development of ABPs for NSPs has significantly advanced research in neutrophil biology, providing powerful tools for investigating protease-based disease mechanisms. Measuring active CatG, NE, or PR3 is far more indicative of ongoing tissue damage than measuring total neutrophil count or total enzyme protein. Biochemical assays often fail to distinguish between active, inactive, or inhibitor-bound NSPs, which can lead to inaccurate conclusions. Assessing active NSP levels is especially beneficial in disease situations where neutrophil- driven tissue damage is significant, such as COPD, ARDS, or IBD. In COPD, excessive NSP activity contributes to extracellular matrix degradation and airway remodeling. In ARDS and cystic fibrosis (CF), NE-mediated lung injury accelerates disease progression (Korkmaz et al., 2010). In inflammatory conditions, such as IBD, conventional methods have been insufficient in correlating total NSP levels with disease severity. However, ABPP has successfully identified active NE exclusively in inflamed colon tissue (Anderson et al., 2019). These findings underscore the importance of activity-based detection in providing more accurate understanding of NSP functions. 1.4.3 Development of ABPs for NSPs ABPs have been successfully applied in distinct cell-based assays to study NSP activity. In flow cytometry-based experiments, ABPs have been successfully applied to detect NSP activity extra- and intracellularly, while SDS-PAGE-based ABPP was used to identify probe- labeled NSPs in cell lysates. Moreover, fluorescence microscopy was applied to detect NSP activity in various cells, offering insights into NSP localization and interaction with host cells. The foundation for developing ABPs for NSPs was established in the late 1970s with the discovery of α-aminoalkylphosphonate diphenyl esters (DAP) as irreversible inhibitors of serine proteases (Oleksyszyn et al., 1979). In 1991, Oleksyszyn and Powers introduced diaryl α-aminophosphonate derivatives, which were among the first tailored ABPs designed to detect and inhibit NSPs. These inhibitors helped define the biochemical characteristics of NSPs, including their substrate preferences and active site specificity, paving the way for later probe development and serving as precursors to modern ABPs. 28 1.4.3.1 Development and structure of ABPs to detect active CatG A significant advancement in detecting active CatG involved modifying an established inhibitor by incorporating a detection tag. One of the earliest CatG-targeting ABPs, Bt-Suc- Ala-Ala-PheP(OPh)2 (DAP22c-biotin), was developed by attaching a biotin moiety to a potent CatG inhibitor (Oleksyszyn and Powers, 1991). However, its sensitivity was limited, requiring at least 20 µg of cell lysate for effective detection of active CatG (Reich et al., 2009). To enhance sensitivity the probe was refined based on the most potent irreversible CatG inhibitor, Suc-Val-Pro-Phe(OPh)2 (SucVPF), by featuring a biotin moiety linked through a long-chain spacer (denoted as “Bt-LC”) to the succinylated peptide (Oleksyszyn and Powers, 1991; Zou et al., 2012). The design was inspired by Oleksyszyn’s 1991 research, incorporating optimal amino acids at the P1, P2, and P3 positions within a peptidyl diphenyl phosphonate structure to enhance CatG specificity. The warhead is an electrophilic phosphonate ester attached to the Phe residue (denoted PheP(OPh)2). This warhead mimics the normal scissile bond peptide bond and, upon binding, the Ser195 nucleophile attacks the phosphorus atom, releasing one phenoxy group and forming a stable covalent phosphoryl- enzyme complex (Burster et al., 2021). This probe Bt-LC-Suc-Val-Pro-Phe(OPh)2 (referred to MARS116-Bt; MARcin Sienszyk) enabled high-sensitivity CatG detection in just a few micrograms of cell lysate. Its application in high-throughput ELISA assays allowed for sub nanomolar detection via biotin-streptavidin signal amplification (Zou et al., 2012). Despite the improvement of MARS116-Bt, subsequent studies revealed that MARS116-Bt was ineffective in detecting intracellular CatG activity due to the non-specific binding of streptavidin to biotin within cells. To overcome this limitation, MARS116-FAM was developed by incorporating 5(6)-carboxyfluorescein conjugated via a 1,5-diaminopentane (DAPe) linker (Schroeder et al., 2020). The fluorescent version allowed for faster handling prior to flow cytometry analysis, as it eliminated the additional staining and washing steps associated with the biotin-avidin detection system. Furthermore, MARS116-FAM provided more direct and efficient visualization of active CatG in SDS-PAGE-based applications and intracellular flow cytometry analysis. However, the ability of MARS116-FAM to bind membrane-bound CatG was not entirely clarified (not published). This was likely due to limitation in detecting cell membrane bond CatG. To address this issue, MARS116-FAM was truncated by a CH2 group at the spacer region, which might reduce steric hindrance and improved probe accessibility. Therefore, the modification resulted in the development of MARS116-FAM ∆C5, a novel variant designed 29 to enhance the detection of membrane-associated active CatG. Notably, a non-reactive analog of MARS116 was also synthesized as a control probe, which lacked the phosphonate warhead and instead contained a carboxylate (Phe-COO- instead of PheP (OPh)2). The ABP is referred to MARS116*-FAM, which is incapable to form a covalent bond with the catalytic center of CatG and serves to assess non-specific binding under experimental conditions (negative control) (Schroeder et al., 2020). 1.4.3.2 Other NSP ABPs By varying the P1 residue, related ABPs have been synthesized to target other NSPs. For example, MARS123 (P1=Leu) and MARS125 (P1=Val) are analogs designed to align with NE or PR3 preferences. Notably, these probes do not label CatG under normal conditions, underscoring the significance of the Phe residue at P1 in MARS116 for CatG specificity. In this study, neither Leu- nor Val-based probes bound CatG unless the substrate selectivity for CatG was altered by allosteric modulation by lactoferrin (LF). Likewise, MARS116 does not react with NE or PR3 under normal conditions (Eipper et al., 2016). In conclusion, the MARS116 amino acid sequence is uniquely designed for CatG to detect selective CatG activity in biological samples, while other NSPs require different ABPs. 1.4.4 Application of MARS116 in flow cytometry assays MARS116 has emerged as a versatile tool for detecting active CatG in biological samples, which allows for subsequent detection of proteolytic active CatG by using flow cytometry and other single-cell techniques including CyTOF (Gärtner et al., 2020). The high specificity of MARS116, combined with irreversible and covalent binding to the catalytic center of CatG, enables robust detection with minimal background interference. Additionally, the probe’s design, incorporating a tripeptide sequence motif and a biotin/fluorophore tag, ensured that even low levels of CatG activity can be detected. In flow cytometry, detection of MARS116-labeled enzymes can be achieved in two distinct approaches: 1. Indirect detection (MARS116-Bt), which requires post-labeling staining with a fluorescent streptavidin-conjugate or an anti-biotin antibody. The selection capacity of MARS116-Bt and its sensitivity proved valuable for studying antigen-presenting cells (APCs) and other immune cell populations. For instance, after labeling cell- surface CatG with MARS116-Bt and subsequently adding streptavidin-FITC to detect 30 proteolytic active CatG by flow cytometry (Penczek and Burster, 2019). Penczek et al. (2016) demonstrated that a distinct subset of human natural killer (NK) cells carry active CatG on their surface. Applying MARS116-Bt in flow cytometry, the authors identified differences in CatG activity between CD56dim versus CD56bright NK cell subsets. These findings expand the function of CatG on NK cells. The authors concluded that “MARS116-Bt is a novel reporter of cell surface CatG activity” that can be used to phenotypically differentiate immune cell subsets. Beyond NK cells, MARS116-Bt has also been used to confirm CatG activity in neutrophil-derived supernatants via ABPP and could be similarly applied in flow cytometry to quantify surface-bound CatG on neutrophils (Eipper et al., 2016). This is particularly useful to identify activated neutrophils that have released CatG onto their surface, a key feature of neutrophil-mediated inflammation. Additionally, mass cytometry (CyTOF) has incorporated an anti-biotin antibody tagged with a rare-earth metal, allowing high- dimensional profiling of active CatG across various immune cell subsets without the need to isolate cells or perform separate assays to determine the proteolytic activity of the protease (Gärtner et al., 2020) 2. Direct detection (MARS116-FAM), where a fluorescence label is conjugated to the ABP, of proteolytic active CatG. This simplified protocol is particularly useful for intracellular CatG detection, where the use of secondary reagents results in high background fluorescence and unspecific binding (Schroeder et al., 2020). Schroeder et al., 2020 applied MARS116-FAM to measure intracellular CatG activity in CD4+ T cells, CD8+ T cells, and regulatory CD4+CD25+ T cells, where CatG could be taken up or being released from these cells during an immune response. Thus, even in non- myeloid cells intracellular CatG activity can be monitored by flow cytometry since MARS116-FAM is also cell-permeable. 1.4.5 Advantages and future perspectives of MARS116-FAM In all of the above examples, MARS116 provides a functional assay to detect CatG activity rather than simply measuring total protein levels. This distinction is important in cases where CatG is present but inhibited by endogenous regulators, such as serpins, or present as a zymogen, which is a powerful feature for studying the regulation of NSPs in both host defense but also in inflammatory diseases (Burster et al., 2021). The development of ABPs for NSPs has evolved from broad-spectrum inhibitors in 1979 to highly selective, real-time imaging probes in the 21st century, which opens opportunities to 31 study NSP activity in specific cell subsets and neutrophil-epithelial cell interactions. Moreover, ABPP can be “multiplexed”, allowing simultaneous detection of different NSPs using selective ABPs harboring different fluorophores or combined with other cell markers. ABPs are being explored beyond classical immunology, with applications in cardiovascular research, chronic inflammatory diseases, and virology, for instance SARS-CoV-2. One can also assess the efficacy of protease inhibitors and convert it into an ABP, a crucial step in drug development, providing a function readout by inhibition of the protease. 32 1.5 Study rationale, aims, objectives, and hypotheses NSPs are well-established effectors of the innate immune system, playing critical roles in pathogen clearance, inflammation, and tissue remodeling. While their presence and activity have been extensively characterized in immune cells, the potential interactions of NSPs with other cells, including epithelial cells remain poorly explored. Given that neutrophils frequently degranulate in close proximity to mucosal barriers, such as the airway epithelium, it is logical to determine whether NSPs bind to epithelial surfaces and remain proteolytically active for functional purposes. Current evidence suggests that NSPs are found on the surfaces of immune cells, where they contribute to intercellular signaling and immune defense. However, very few studies have investigated their localization and activity in epithelial cells. If epithelial cells are capable of binding and retaining active NSPs, this could represent an additional, yet underscored, mechanism of host defense. The lack of sensitive detection tools has limited such investigation, making it unclear whether protease localization to epithelial surfaces is an incidental event or a regulated physiological process. ABPP complements existing approaches by offering direct, near-real-time detection of active proteases. Unlike standard antibody-based techniques, ABPP specifically labels enzymatically active NSPs. This functional approach allows for a more precise characterization of protease activity in complex biological systems, making it an ideal tool for studying enzymatic interactions with epithelial cells. Using ABPs, it will be investigated whether NSPs bind to lung epithelial cells, remain enzymatically active, and potentially contribute to immune defense. In order to systematically explore the above inquiries, this study is structured around the indicated hypotheses, each aligned with the corresponding objectives, experimental paradigms, and a detailed step-by-step methodology. The investigation commences with a biochemical validation of the ABPs utilizing SDS-PAGE, followed by cell-based assays employing flow cytometry and confocal microscopy as well as employing protease inhibitors. Hypothesis: ABPs can specifically and sensitively detect active proteases ABPP enables the detection of active proteases, distinguishing them from inactive or inhibited forms. This hypothesis proposes that fluorescent ABPs will selectively label only catalytically 33 active NSPs and that inhibitor pre-treatment will abolish probe labeling, confirming the specificity of the approach. Aim: Validate the specificity and sensitivity of ABPs in detecting NSP activity. Objectives: ● SDS-PAGE experiments will be performed using known NSP inhibitors (e.g., CatGinh, SucVPF, and sivelestat) in competitive inhibition assays to assess whether the binding of the ABP to NSPs is specific. ● Flow cytometry experiments further evaluate specificity by comparing ABPs with and without the respective inhibitor. ● Confocal microscopy will determine the localization of enzymatic activity with ABP signals. Key experiments: Titration of CatGinh to inhibit CatG or sivelestat to inhibit NE to demonstrate dose-dependency of reduction in MARS116-FAM or VPV-FAM labeling using SDS-PAGE and confocal microscopy. Hypothesis: Neutrophil proteases CatG and NE might bind to lung epithelial cell surfaces in an active manner Although NSPs are traditionally associated with immune cells, it is unclear whether lung epithelial cells can retain active NSPs. This might suggest that lung epithelial cells acquire NSPs from activated neutrophils, which secrete NSPs, potentially enhance local immune defense. Aim: To investigate whether CatG or NE can bind to epithelial cell surfaces and retain proteolytically active. Objectives: ● To confirm probe labeling efficiency, purified CatG and NE will be assessed in SDS- PAGE-based ABPP. ● To evaluate whether exogenous NSPs bind to lung epithelial cells in vitro, flow cytometry-based ABPP will be used. ● To visualize NSP binding, confocal microscopy will be employed, investigating spatial localization of active NSPs on epithelial surfaces in a time-dependent fashion. 34 Key experiment: Time-course imaging (5-60 min) to track protease localization. Hypothesis: Cellular conditions, such as nutrient availability (e.g., serum starvation), could influence protease binding/activity on epithelial surfaces. Serum contains protease inhibitors that may regulate NSP retention on epithelial cells. Therefore, it is hypothesized that starvation conditions (i.e., serum depletion) would increase detectible NSP activity, whereas serum-rich conditions might suppress NSPs. Aim: Determine the impact of serum starvation on NSP binding and activity Objectives: ● To compare CatG binding under serum-free and serum-rich conditions, confocal microscopy starvation experiments will be performed using MARS116-FAM. Key experiment: Parallel experiments with or without serum to reveal CatG activity. While these initial hypotheses focus on the fundamental interactions between NSPs and epithelial cells, the study also seeks to expand toward ABPP methodologies and therapeutic potentials. Specifically, the aim is to address critical gaps in the application of ABPs for NSP detection, as well as explore potential therapeutic strategies to modulate NSP activity. To this end, additional hypotheses are postulated: Hypothesis: MARS116-FAM ∆C5 might show superior sensitivity for membrane bound CatG compared to MARS116-FAM. Initial studies suggested that MARS116-FAM struggled to detect membrane-bound CatG, potentially due to steric hindrance. To overcome this, a truncated variant (MARS116-FAM ∆C5) was developed. The hypothesis is as follows, MARS116-FAM ∆C5 could improve membrane labeling by reducing steric hindrance. Aim: Compare the performance of MARS116-FAM and MARS116-FAM ∆C5. Objectives: ● SDS-PAGE experiments will be conducted and compare probe sensitivity for detecting active CatG. 35 ● Both ABPs will be tested by flow cytometry assays on epithelial cells to determine whether the shorter linker improves labeling. ● Confocal microscopy time-course experiments will be applied to visually both probes. Key experiments: Direct comparison of fluorescence intensity and localization between probes. Hypothesis: VPV-FAM might label NE and PR3 with shared specificity The ABP FAM-DAPe-Suc-Val-Pro-ValP(OPh)2 (VPV-FAM) was synthesized based on the substrate preferences of NE and PR3. Both NE and PR3 prefer small aliphatic residues at P1 position, suggesting overlapping substrate specificity. Here it is hypothesized that VPV-FAM would effectively label both NE and PR3. Ultimately, the objective is to test whether the ABP can detect membrane-bound as well as intracellular NE in lung epithelial cell lines. Aim: Assess the specificity of VPV-FAM for NE and PR3. Objectives: ● SDS-PAGE experiments will be performed to compare VPV-FAM labeling for NE and PR3 to determine sensitivity and selectivity. ● Inhibitor competition assays will be applied by using sivelestat to assess specificity. ● To visualize NE binding, confocal microscopy will be employed, investigating localization of active NE in lung epithelial cell lines as well as at the cell surface. Hypothesis: Camostat mesylate might inhibit NSPs in a dose-dependent manner. Given the established efficacy of camostat in inhibiting the serine protease TMPRSS2, which plays a critical role in SARS-CoV-2 viral entry by cleaving its Spike protein, here it is hypothesized that camostat might similarly inhibit NSPs, particularly CatG, NE, and PR3, like TMPRSS2. Structural similarities in the active-site regions between TMPRSS2 and NSPs, including the conserved catalytic triad (His, Asp, and Ser), suggest potential susceptibility to camostat inhibition; therefore, testing camostat’s activity against NSPs is scientifically justified. The experiments will be performed to explore the possibility to extend the therapeutic application of camostat beyond viral inhibition to modulating inflammatory responses, thereby providing dual benefits in the context of viral infections characterized by exacerbated inflammation. Hence, inhibiting NSPs could yield therapeutic benefits by both preventing viral propagation and mitigating inflammation. 36 Aim: To test camostat’s inhibitory capacity towards CatG, NE, and PR3. Objectives: ● SDS-PAGE-based ABPP will be used to test whether camostat inhibits CatG, NE, or PR3. ● Molecular docking will be performed to model camostat-NSP interactions, assessing binding affinity and inhibition mechanism. By addressing these hypotheses, the study aims to fill critical gaps in our understanding of NSP interactions with epithelial cells while advancing the development and validation of novel tools and potential therapeutic strategies for studying and modulating NSP activity. Demonstrating that CatG and NE can bind epithelial surfaces in a proteolytically active manner would provide insights into the novel mechanism of host defense - one in which epithelial cells are “armed” by NSPs to directly combat pathogens at the mucosal barrier. Additionally, validating the use of ABPP will support its broader utility in studying enzyme activity on the cell surface. The outcome from this work will provide insights into the roles of NSPs and potential applications in both basic science and clinical research. 37 Chapter 2 Materials and Methods 2.1 Materials 2.1.1 Chemicals and consumables Chemicals were sourced from ThermoFisher Scientific (Carlsbad, CA, USA), Sigma-Aldrich (St. Louis, MO, USA), and Bio-Rad Laboratories (Hercules, CA, USA), unless specified otherwise. Purified CatG, NE, and PR3 were obtained from either Athens Research and Technology (Athens, GA, USA) or Abcam (Cambridge, MA, USA). Consumables, including pipette tips, reaction tubes, centrifuge tubes, and cell culture flasks were supplied by Corning (Corning, NY, USA), Techno Plastic Products AG (Trasadingen, Switzerland), and VWR (Radnor, PA, USA). 2.1.2 Activity-based probes The ABPs were kindly provided by Professor Dr. Marcin Sienczyk and Professor Dr. Renata Grzywa (Division of Medicinal Chemistry and Microbiology, Faculty of Chemistry, Wroclaw University of Technology, Wroclaw, Poland). Sunflower trypsin inhibitor (SFTI)-22 was kindly provided by Professor Dr. David Craik and Professor Dr. Simon de Veer (Institute of Molecular Biosciences, University of Queensland, Queensland, Australia). To detect active NSPs, ABPs were designed to covalently label the catalytic serine residue of target proteases. These probes consist of (1) a nucleophilic phosphonate warhead, (2) a recognition sequence (mimics the tripeptide preference of the target protease), and (3) a FAM tag. The ABPs used for this thesis: ● MARS116-FAM (FAM-(CH2)5-Suc-Val-Pro-PheP(OPh)2) - targets CatG with the recognition sequence Val-Pro-Phe (SucVPF). ● MARS116-FAM ΔC5 (FAM-(CH2)4-Suc-Val-Pro-PheP(OPh)2) - a variant of MARS116-FAM with a shorter spacer (deletion of CH2 at position 5; ΔC5) that was designed to possibly overcome the limitation of the detection capacity of MARS116- FAM’s positioning of the FAM tag relative to the active site. ● VPV-FAM (FAM-DAPe-Suc-Val-Pro-ValP(OPh)2) - targets elastase-like proteases such as NE and PR3 with the Val-Pro-Val tripeptide sequence. VPV-FAM was expected to irreversibly label active NE and PR3 in our assays due to their structural homology and shared substrate preferences. 38 ● MARS116*-FAM - A control analog without the reactive phosphonate, was used to exclude nonspecific binding. 2.1.3 Chemical inhibitors To confirm probe specificity, protease inhibition was performed using: ● SucVPF (Suc-Val-Pro-PheP(OPh)2): a gold-standard irreversible CatG inhibitor that shares the same tripeptide recognition sequence and phosphonate warhead as MARS116-FAM. This compound was synthesized based on Oleksyszyn’s peptidyl phosphonate design (Oleksyszyn and Powers, 1991). ● CatGinh (Cathepsin G Inhibitor I; Calbiochem, Merck Chemicals GmbH, Schwalbach, Germany): a potent, selective, reversible inhibitor of CatG (Calbiochem, Cat. No. 219072). This non-peptide inhibitor has an IC=53 nM for CatG and exhibits minimal off-target activity against other serine proteases (nearly no inhibition of PR3 or NE at >100 µM (Greco et al., 2002). ● Sivelestat (ONO-5046; Tocris Bioscience, Bristol, UK) - a synthetic, low-molecular- weight competitive inhibitor of NE and PR3, used to treat acute lung injury (Rajagopal et al., 2024). To test whether NSPs can be inhibited by a broad-spectrum serine protease inhibitor, camostat mesylate (SML0057, Sigma-Aldrich, St. Louis, MO, USA) was used. 2.1.4 Laboratory equipment Autoclave Systec DX-23,Systec GmbH & Co. KG, (Linden, Germany) Balances LA203E Precision Balances, Mettler Toledo, (Columbus, OH, USA) Centrifuges Microfuge™ 22R, Beckman Coulter (Indianapolis, IN, USA) 5702, Eppendorf (Hamburg, Germany) Sorvall ST1 Plus, ThermoFisher Scientific (Carlsbad, CA, USA) Confocal microscope ZEISS LSM 780, Carl Zeiss AG (Jena, Germany) 39 Gel electrophoresis chamber XCell SureLock™ Mini-Cell, ThermoFisher Scientific (Carlsbad, CA,USA) Ice generator Scotsman SCE170A-1C (Vernon Hills, IL, USA) Incubator HERAcell 150i, ThermoFisher Scientific (Carlsbad, CA, USA) Inverted microscope Zeiss PrimoVert, Carl Zeiss Microscopy (White Plains, NY) Laminar flow Safemate 1.5 eco, EuroClone (Pera, Italy) pH meter STAR A211 Star A211 pH Benchtop Meter, ThermoFisher Scientific (Carlsbad, CA, USA) Power supply for gel electrophoresis PS300B, Hoefer (Bridgewater, MA, USA) Shakers MaxQ 2000, ThemoFisher Scientific (Carlsbad, CA, USA) AP9749, DLAB (Beijing, China) Thermoblock Thermomixer C Model 5382, Eppendorf, (Hamburg Germany) Water bath Precision GP 20, ThermoFisher Scientific (Carlsbad, CA, USA) Water purifier Smart2Pure UV, ThermoFisher Scientific (Carlsbad, CA, USA) 2.2 Methods 2.2.1 Sodium dodecyl sulfate - polyacrylamide gel electrophoresis (SDS-PAGE) coupled with activity-based protein profiling (ABPP) ABPP coupled with SDS-PAGE was employed to visualize and confirm proteolytic activity and specificity of NSPs. The protein size and in-gel fluorescence intensity was assessed by ABPs in order to detect the proteolytic activity of NSPs by SDS-PAGE. This method is referred to as activity-based protein profiling (ABPP). Prior to SDS-PAGE, purified proteases, CatG, NE, or PR3 (Athens Research and Technology, Athens, GA, USA or Abcam, Cambridge, MA, USA) were labeled with fluorescent activity-based probes (MARS116- FAM, MARS116-FAM ∆C5, or VPV-FAM) for 40 minutes at room temperature (RT) 40 protected from light. The probe-enzyme complex remains stable even after denaturation to be suitable for SDS-PAGE. The SDS-PAGE system, consisting of a 12% acrylamide resolving gel and a stacking gel, prepared in standard 1 mm cassettes (novex, ThermoFisher Scientific, Carlsbad, CA, USA). Resolving gels were poured into cassettes, overlaid with 96% EtOH and left to polymerize for approximately 20 minutes to polymerize at RT. After polymerization, EtOH was removed, and the stacking gel was poured over the resolving gel. A 12-well comb was inserted, and polymerization proceeded for another 20 min at RT. Gels were stored at 4ºC in a plastic bag, covered with moist tissue paper to prevent drying. Component Resolving gel Stacking gel ddH2O* 6.6 ml 6.8 ml 30% Acrylamide* 8 ml 1.66 ml 1M Tris HCl, pH 6.8 - 1.26 ml 1.5 M Tris HCl, pH 8.8 5 ml - 10% SDS 200 µl 100 µl TEMED* 8 µl 10 µl 10% APS* 200 µl 100 µl Table 1. Components for 12% and 5% SDS polyacrylamide gels. *Double distilled water; ddH2O, N,N,N′,N′- Tetramethylethylenediamine; TEMED (Thermo Fisher Scientific, Carlsbad, CA, USA), ammonium persulfate; APS. 30% bis-acrylamide (A3574, Sigma-Aldrich, St. Louis, MO, USA), Tris/Base (BP152, ThermoFisher Scientific, Carlsbad, CA, USA). In general, SDS was used as a protein denaturation agent, while 4x loading buffer acted as a reducing agent to reduce disulfide bonds. Samples were boiled at 95ºC for 5 min before loading onto a gel, followed by electrophoresis, performed at 100V for ~20 min (stacking phase) and 220V for ~45 min (separation phase). To confirm proper protein loading, Precision Plus Protein Dual Color Standards (Bio-Rad, Hercules, CA, USA) was included as molecular weight (MW) markers, providing reference bands at known MWs (10-250 kDa). The probe-enzyme complex remains stable in the SDS-PAGE gel and can be detected by fluorescence imaging using ChemiDoc Touch Imaging System (Bio-Rad Laboratories, Inc., Hercules, CA, USA) using the Pro-Emerald 488 filter channel. To verify equal protein loading across lanes (loading control), gels were stained with Coomassie Brilliant Blue and placed on a shaker for 15 minutes. Excess stain was removed 41 through two consecutive destaining steps (~20 min each) in destaining solution until the protein bands were visualized. The stained gels were imaged using the white light mode on the ChemiDoc Touch Imaging System. Finally, fluorescence and protein loading images were generated using ImageJ (Schindelin et al., 2012; NIH, Bethesda, MD, USA; DOI:10.1038/nmeth.2019). All SDS-PAGE experiments were performed at least two times and indicated in the figure legend. 2.2.2 Cell culture and maintenance The A549 and H1299 lung epithelial cell lines were chosen as model systems due to their physiological resemblance to alveolar type II cells, primary targets of viral pathogens, including SARS-CoV-2. Table 2. Cell lines. The H1299 cell line was kindly provided by Prof. Dr. Dos Sarbassov and A549 cell line was a gift from Prof. Dr. Gonzalo Hortelano (Department of Biology, School of Sciences and Humanities, Nazarbayev University). All cell culture methods described were carried out in a sterile laminar flow hood (Safemate 1.5 eco, EuroClone, Pera, Italy). Cells were cultured in Roswell Park Memorial Institute Running buffer 250 mM 1.92 M 1% (w/v) Tris base Glycine SDS Coomassie brilliant blue staining solution 0.25% 45% 10% Coomassie brilliant blue Methanol Acetic acid Destaining solution 45% 10% Methanol Acetic acid Cell line ATCC no. Origin A549 CCL-185 American Type Culture Collection (ATCC, Manassas, VA, USA); adenocarcinoma human alveolar epithelial cell line; alveolar type II cells NCI-H1299 CRL-5803 ATCC; human epithelial-like, non-small cell lung carcinoma cell lined derived from the lymph node 42 (RPMI)-1640 medium (R8755-1L, Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (FBS, 2631713RP, Gibco, Grand Island, NY, USA) unless mentioned otherwise; 1% Penicillin-Streptomycin (pen-strep, 210496, Gibco, Grand Island, NY, USA) to form a complete growth medium. Medium and other solutions were autoclaved or sterile filtered using 0.22 µm pore size syringe filter or a vacuum filter system (Corning, NY, USA) . Cells were maintained in a humidified CO2 incubator (HERAcell 150i, Thermo Fisher Scientific, Carlsbad, CA, USA) at 37 ºC with 5% CO2. The medium was changed every 2 to 3 days. Multisizer 4e Coulter Counter (Beckman Coulter, Indianapolis, IN, USA) was used to identify cell density, while cell viability was determined on Countess 3 (Thermo Fisher Scientific, Carlsbad, CA, USA) after mixing trypan blue solution; cell suspension 1:1 ratio. Trypan blue enters the perforated membranes of dead cells and allows for discrimination living cells by the machine automatically. A549 and H1299 cells were propagated until they reached 80-90% confluency. The passaging procedure included culture medium aspiration followed by washing with pre-warmed phosphate-buffered saline (PBS), pH 7.4 (RNBL5991, Sigma-Aldrich, St. Louis, MO, USA). 1 ml of 0.5% Trypsin-EDTA (2537762, Gibco, Grand Island, NY, USA) was added to detach cells and placed into the CO2 incubator for 3 min at 37ºC (or until visual detachment was observed under an inverted microscope (Zeiss PrimoVert, Carl Zeiss Microscopy, White Plains, NY). The detached cells were resuspended in 8 ml of complete growth medium for trypsin neutralization. The cell suspension was transferred to a 15 ml conical tube (Corning, Charlotte, NC, USA) and centrifuged for 5 min at 300 x g (Centrifuge 5702, Eppendorf, Hamburg, Germany) at RT. The supernatant was aspirated, and the cell pellet was resuspended in fresh complete growth medium. Cells were reseeded at an appropriate density for further culturing. For long-term storage, cells were cryopreserved in a freezing medium consisting of 80% complete medium, 10% FBS, and 10% (v/v) dimethyl sulfoxide (DMSO, D8418, Sigma- Aldrich, St. Louis, MO, USA). Cells were harvested at the exponential growth phase, centrifuged at 300 x g for 5 min, and resuspended in the pre-cooled freezing medium at a density of 1 x 106 cells/ml. The cell suspension was aliquoted into cryovials, which were then stored at -80ºC in a freezer or liquid N2. For thawing, cryovials were rapidly warmed in a 37ºC water bath with gentle agitation until only a small ice pellet remained. The thawed suspension was immediately transferred to a 15 43 ml conical tube containing 10 ml pre-warmed complete medium to dilute DMSO. Cells were centrifuged at 300 x g for 5 min, the supernatant was aspirated, and the pellet was resuspended in a “fresh” complete medium. The recovered cells were plated in vented T25 culture flasks (Corning, Charlotte, NC, USA or TPP 90025, Techno Plastic Products AG, Trasadingen, Switzerland) and incubated under standard conditions, with a medium range of 24 hours for further experimental tasks. Authentication of A549 and H1299 cell lines was confirmed by short tandem repeat (STR) profiling. Detailed STR results are presented in Appendix A. It should be noted that initial experiments were performed with cells believed to be A549 cell line, however outsourced STR analysis revealed that the cell stock corresponded to H1299 cells. Upon this discovery, subsequent experiments were conducted using authenticated A549 cells and all key experiments were reproduced to ensure consistency. Notably, H1299 (misidentified) and authenticated A549 yielded comparable results, demonstrating that these two lung epithelial cell lines serve as similar models for the study. The reproducibility of results across two cell lines strengthens the generalizability of the conclusions made. 2.2.3 Measurement of membrane-associated CatG activity by flow cytometry Flow cytometry provided quantitative assessment of CatG bound to the cell surface, allowing precise measurement of enzyme activity at a single-cell resolution. Sample preparation A549 or H1299 cells were cultured in RPMI-1640 supplemented with 10% FBS and 1% pen- strep until 90% confluency, trypsinized, and resuspended in standard medium to neutralize trypsin, followed by centrifugation at 300 x g for 5 min. To remove residual media, cells were washed in PBS and subsequently resuspended in PBS supplemented with 1% FBS (FACS buffer) to minimize non-specific interactions. The cell suspension was adjusted to a final concentration of 5 x 106 cells/ml. Purified CatG was added to achieve a final concentration 10 µg/ml and incubated for 1 hour at RT. Following this, the desired number of cells per experiment (1 x 105 cells/sample was calculated using the following formula: 𝑥 = 𝑡𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑐𝑒𝑙𝑙𝑠 𝑑𝑒𝑠𝑖𝑟𝑒𝑑 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑐𝑒𝑙𝑙𝑠 (𝑑𝑖𝑣𝑖𝑑𝑖𝑛𝑔 𝑓𝑎𝑐𝑡𝑜𝑟) 44 𝑦 = 𝑣𝑜𝑙𝑢𝑚𝑒 𝑜𝑓 𝑐𝑒𝑙𝑙𝑠 𝑥 (𝑣𝑜𝑙𝑢𝑚𝑒 𝑜𝑓 𝑐𝑒𝑙𝑙𝑠 𝑛𝑒𝑒𝑑𝑒𝑑 𝑝𝑒𝑟 𝑠𝑎𝑚𝑝𝑙𝑒) The appropriate volume of cells was transferred into 2 ml round-bottom microcentrifuge tubes (Eppendorf, Hamburg, Germany) and gently vortexed. Labeling of CatG with probes Following the 15-minute pre-incubation with CatGinh [50µM], increasing concentrations of MARS116-FAM or MARS116-FAM ∆C5 (ranging from 1.25 to 10 µM) were added simultaneously to both CatG-CatGinh complex and CatG-only samples (without inhibitor) and incubated for an additional 40 minutes at RT, allowing covalent binding of the probe to the enzyme active site. An unstained sample was included as a negative control. All incubation steps were performed under light-protected conditions to maintain fluorophore stability. Following incubation, cells were washed twice in FACS buffer (PBS pH 7.4 + 1% FBS) to remove unbound probe and residual complexes. Finally, cells were resuspended in 400 µl of FACS buffer and stained with propidium iodide (PI; 0.5 µg/ml, Sony Biotechnology, San Jose, CA, USA)) for live/dead cell discrimination. Data acquisition and analysis Samples were analyzed on Attune NxT Flow Cytometer (ThermoFisher Scientific, Carlsbad, CA, USA) with fluorescence excitation at 488 nm for FAM-conjugated probes and 561 nm for PI. A minimum of 30,000 gated events were recorded per sample to ensure statistical robustness. Data was analyzed using FlowJoTM v10.8.1 Software (BD Life Sciences, Ashland, OR, USA) using gating strategy (shown below) to minimize background noise and exclude non-relevant events. Further statistical analysis was performed as described in Section 2.4. Gating strategy Forward Scatter Area (FSC-A) vs. Side Scatter Area (SSC-A) plots were used to identify and gate the main cell population of interest, ensuring the exclusion of debris and non-cellular particles (Fig 3A).To ensure that only single cells were analyzed, a secondary gate was applied to distinguish singlets from aggregates or doublets by FS-Height (FSC-H) vs. FSC-A plot. The linear single-cell population was selected (Fig 3B). The single-cell population was further gated based on PI staining, with PI-negative cells considered viable (Fig 3C). 45 Fig 3. Flow cytometry gating strategy for the detection of surface-bound CatG. A549 or H1299 single-cell suspensions were prepared and stained according to the described protocol. Flow cytometry was used to analyze membrane-bound CatG activity using MARS116-FAM or MARS116-FAM ∆C5. (A) Cells were first gated based on forward and side scatter parameters to exclude debris and identify the cell population of interest. (B) A single-cell gate to eliminate doublets. (C) Propidium iodide (PI)-negative cells were gated. (D) Unstained cells (FITC-A-) appear in grey, ABP-stained cells are represented in dark grey. (E-F) ABP fluorescence; data from all tested ABP concentrations [1.25-10 µM] were combined to evaluate the dose-dependent labeling of CatG on the cell surface. All values are reported as median fluorescent intensities (MFI). MARS116-FAM [10 µM] was used as the single-stained control to define fluorescence detection parameters and ensure proper gating of probe-positive cells (Fig 3D). Spectral compensation was performed using single-stained controls for all probes as well as PI to correct for spectral overlap between fluorescence channels to prevent a false-positive signal. To assess overall trends across probe concentrations and to visualize the general distribution of fluorescence intensity, all MARS116-FAM and MARS116-FAM ∆C5 samples across concentrations [1.25 µM-10 µM] were concatenated in FlowJoTM (Fig 3E and F). This allowed directly quantify fluorescence intensity. 2.2.4 Confocal microcopy Confocal microscopy was employed to directly visualize and localize protease activity at the cellular membrane, providing spatial and temporal resolution of enzyme-cell interactions. 46 A549 and H1299 human lung epithelial cell lines were cultured in standard conditions (RPMI- 1640 supplemented with 10% FBS and 1% pen-strep until 90% confluency) in glass-bottom µ-slides (Ibidi GmbH, Gräfelfing, Germany), and 24 hours prior to the experiment, the medium was replaced with serum-free media to minimize interference from serum-derived protease inhibitors. In general, CatG and NE at 10 µg/ml concentrations, were pre-labeled with their respective probes to enable visualization of active enzyme binding. CatG was labeled with MARS116-FAM or its truncated variant, MARS116-FAM ∆C5, while NE was labeled with VPV-FAM at indicated concentrations. Labeling was performed in serum-free medium for 40 minutes at room temperature, with precautions taken against fluorophore degradation (protected from light). Following labeling, the protease-probe complexes were transferred to the cultured cells. Cells were incubated with the complexes for specified time point to allow interaction with the cell membrane. Post-incubation, cell were fixed in ice-cold 99% methanol (Sigma-Aldrich, St. Louis, MO, USA) for 5 minutes to preserve cellular architecture and protease localization. Fixed cells were washed twice with PBS, pH 7.4 to remove residual reagents. To visualize nuclei, cells were stained with Hoechst 33342 (Ready Probes™ Reagent, ThermoFisher Scientific, Carlsbad, CA, USA) for 5 minutes before imaging using ZEISS LSM 780 confocal microscope (Carl Zeiss AG, Jena, Germany). Hoechst signal was detected via DAPI channel (excitation/emission: 405/461 nm) and fluorescent ABPs were imaged using the FITC filter (excitation/emission: 488/520 nm). To prevent cross-channel bleed- through, cells were imaged sequentially starting with the nuclear staining (Hoechst). The pinhole was set to 1 AU. Brightfield pictures were captured by PMT trans detector. Images were saved using ZEN microscopy software (version 14.0.23.201, Carl Zeiss AG) and post- processed with FIJI/ImageJ for background subtraction, contrast adjustment, and colocalization analysis. For each slide at least 50 cells were analyzed. For CatG, MARS116*-FAM, a probe lacking the phosphonate warhead, was used as a negative control to exclude non-specific binding. Additionally, cells without probes were imaged to assess and exclude autofluorescence. For specificity, CatGinh [50 µM] and sivelestat [50µM] were added to the negative control samples and incubated for 15 minutes at RT prior the addition of the probes. 47 Probe titration and protease labeling Probe concentrations were systematically optimized to balance signal specificity and cellular visibility. For all proteases (CatG and NE) the following probe concentrations were tested: 0.5, 1, 2.5, 5, and 10 µM. CatG and NE were pre-labeled by incubating with probes in serum- free medium for 40 min at RT. Labeled proteases were added to cells at specified concentrations and incubated for 1 hour at RT protected from light. Note on pre-labeling strategy: In cell-based experiments, NSPs were pre-labeled with ABPs prior to their addition to epithelial cells. This strategy was adopted to ensure consistent probe binding and accurate detection of protease localization. Post-incubation labeling was avoided as it could introduce variability due to enzyme internalization, dissociation, or inactivation during the incubation period. Pre-labeling ensured that only catalytically active enzymes were visualized at the point of cell interaction. Importantly, comparable results were obtained in control experiments where labeling was performed after incubation (data not shown), supporting the stability and reliability of NSP-cell surface interactions. Inhibitor titration assay H1299 cells were incubated with CatG at a concentration of 10 µg/ml. Prior to confocal microscopy imaging, the recombinant CatG was preincubated with increasing concentrations of the specific inhibitor CatGinh [10, 25, and 50 µM] in serum-free media for 15 minutes at room temperature. Following preincubation, MARS116-FAM or MARS116-FAM ∆C5 were added at a concentration of 0.5 µM for additional 40 minutes protected from light. Finally, the CatG-inhibitor-ABP mixture was added to the cells for 1 hour. To ensure specificity, control samples included cells treated with CatG without the inhibitor. The 10 µg/ml concentration of CatG and NE was chosen to reflect the elevated proteases levels seen in inflammatory lung diseases, such as CF, COPD, and bronchiectasis. Clinical studies show NE levels up to 10 µM (~290 µg/ml) and CatG up to 1 µM (~27 µg/ml) in bronchoalveolar lavage fluid, with sputum levels exceeding 10 µg/ml in bronchiectasis patients (Taggart et al., 2005; Korkmaz et al., 2010; Nguyen-Ho et al. 2025). While higher than the 0.005-0.01 µg/ml in healthy individuals, the concentration of 10 µg/ml for both CatG and NE provides a standardized level aligning with the increased protease burden and is suitable for modeling disease-relevant interactions with lung epithelial cells. Beyond disease 48 modeling, this concentration ensures reproducible detection of protease binding and activity in our assays. 2.3 Molecular docking methodology Molecular docking was performed to predict binding modes and interactions of inhibitors with NSP active sites, supporting experimental validations and mechanistic insights. Further, validation of the docking procedure ensured reliability of docking predictions by confirming that the docking procedure could accurately reproduce known inhibitor-binding modes. Molecular docking was performed using AutoDock4 (v4.2.6) for macOS X (Scripps Research, AutoDock, https://autodock.scripps.edu/download-autodock4/, accessed on June 3, 2020), with AutoDock Tools for receptor/ligand preparation. AutoDock4 was selected for the purposes of this study because of its established accuracy, widespread adoption in academic research, and compatibility with covalent and flexible ligand docking (Morris et al., 2009). Target protein structures for human cathepsin G (CatG, PDB ID: 1CGH), neutrophil elastase (NE, PDB ID: 1B0F), proteinase 3 (PR3, PDB ID: 1FUJ), and TMPRSS2 (PDB ID: 7MEQ) were obtained from the Protein Data Bank (PDB, https://rcsb.org). All heteroatoms were removed, and polar hydrogen and partial charges (Kollman/Gasteiger) added to prepare rigid receptor PDBQT files. Ligand structures (camostat mesylate, 4-guanidinobenzoic acid, CatGinh, and SucVPF) were taken from PubChem/BRENDA in SMILES or mol formats, energy-minimized (MMFF94 force field generated in Avogadro) and converted to PDBQT in OpenBabel (Appendix Table A1) (Assylbekova et al., 2021). 2.3.1 Molecular docking parameters Docking simulations employed the Lamarckian genetic algorithm (LGA) with 50 independent runs per ligand and an enhanced population size (ga_pop-size 300) to sample conformations. The LGA allows efficient sampling of ligand conformations while accommodating flexible torsions, making it well-suited for small inhibitors studied for the purposes of this thesis (GhorbanDadrass et al., 2004). A grid box of 40x40x40Å (0.5 Å spacing) centered on each enzyme’s active site was used to encompass the entire catalytic pocket. Fifty ligand poses were generated for each protein-ligand pair, and the lowest-energy binding pose (after clustering at 2.0 Å RMSD) was selected for analysis. The AutoDock scoring function provided an estimated free binding energy (ΔG, kcal/mol) and inhibition constant (Ki) for each pose. Protein-ligand contacts (hydrogen bonds, hydrophobic contacts, salt bridges, pi- stacking interactions) were characterized using Protein-Ligand Interaction Profiler (PLIP, 49 https://plip-tool.biotec.tu-dresden.de/plip-web/plip/index) tool and visualized in the PyMOL Molecular Graphics System (Version 3.0 Schrödinger, LLC, New York, NY, USA) (Assylbekova et al., 2021). 2.3.2 Validation of the docking protocol While more recent commercial docking programs such as Glide (Schrödinger Suite) and Genetic optimization and for ligand docking (GOLD; Cambridge crystallographic data center, CCDC) offer faster and more user-friendly interfaces, AutoDock4 was chosen for its open- source availability and pre-validation protocols using crystallographic ligands (Morris et al., 2009). Docking protocol validation was performed by redocking known inhibitors co- crystallized with each protease. According to the results, AutoDock4 could successfully predict the bioactive conformation. The inhibitor SucVPF from the crystallized SucVPF-CatG complex (PDB 1CGH) was redocked into CatG, while 4-guanidinobenzoic acid (GBA), the active metabolite of camostat, from TMPRSS2-inhibitor structure (PBD 7MEQ) was redocked into TMPRSS2. Common molecular interactions between the