Method and tools for the determination of proteins and of derivatives thereof in cells

Abstract

Method for the determination of at least one structural and / or physico-chemical property of at least one protein or peptide or of properties thereof, said at least one protein or peptide contained in a complex mixture of further proteins and / or other biomolecules in at least one cell, comprising the following steps: 1. delivery of a protease into said cell and limited proteolysis of the complex mixture, followed by cell extraction and / or cell lysis leading to a fragment sample; 2. denaturation of the fragment sample to a denaturated fragment sample; 3. optional complete fragmentation of the denaturated fragment sample in a digestion step to a completely fragmented sample; 4. analytical analysis of the fragmented sample or the completely fragmented sample for the determination of said at least one protein or peptide or of properties thereof.

Description

[0001] TITLE

[0002] METHOD AND TOOLS FOR THE DETERMINATION OF PROTEINS AND OF DERIVATIVES THEREOF IN CELLS

[0003] TECHNICAL FIELD

[0004] The present invention relates to methods for the determination of properties, in particular at least one structural and / or physico-chemical property, in particular conformation and conformational changes, of proteins and of derivatives thereof, in their native biological context, in particular using limited proteolysis e.g. combined with selected reaction monitoring, data-dependent acquisition (DDA), data-independent acquisition (DIA), including Sequential Windowed Acquisition of All Theoretical Fragment Ion Mass Spectra (SWATH) methods and the like.

[0005] PRIOR ART

[0006] A protein’s function is determined by its structure, which changes dynamically to adapt to cellular perturbations. Protein structures change because of post-translational modifications, by interacting with other proteins or molecules, or in response to physical properties like temperature.

[0007] Proteins serve a multitude of functions in the cell, regulating cell growth, differentiation, and survival. A protein’s function is determined by its structure, which may change depending on the cellular context and on cellular perturbations. Changes in protein structure can be subtle, like in an enzyme binding a small regulatory molecule. Other proteins undergo complete remodeling and even form aggregates like in Alzheimer’s or Parkinson’s disease. To understand protein function in health and disease, it is important to study protein structures and detect structural alterations upon various perturbations (e.g., stress, disease, or treatment with drugs). Protein structures are highly dynamic and adapt to their environment and interaction partners.

[0008] Available nuclear magnetic resonance (NMR) spectroscopy-based methods, and Forster resonance energy transfer (FRET) techniques provide information on the dynamics of protein structures inside intact cells, but only one or a few proteins can be studied at a time. On the other hand, structural proteomics techniques like limited proteolysis coupled mass spectrometry (LiP-MS) are used to study the dynamics of thousands of protein structures simultaneously. Cryo-electron microscopy (cryoEM) techniques can solve protein structures in cells and tissues but only proteins fulfilling certain size constraints can be analyzed. It only allows for a limited degree of multiplexing and is therefore not globally applicable in an unbiased fashion. Finally, issues of resolution and low throughput hamper detection of protein conformational changes.

[0009] WO-A-2014082733 discloses said limited proteolysis (LiP-MS) protocol, i.e. a method for the detection of the conformational state of a protein contained in a complex mixture of further proteins and / or other biomolecules, in particular in a complex native biological matrix, as well as assays for such a method. The method comprises, if needed after an extraction and / or lysis step, the following steps: 1. Limited proteolysis of the complex mixture under a condition where the protein is in the conformational state to be detected leading to a first fragment sample; 2. Denaturation of the first fragment sample to a denaturated first fragment sample; 3. Complete fragmentation of the denaturated first fragment sample in a digestion step to a completely fragmented sample; 4. Analytical analysis of the completely fragmented sample for the determination of fragments characteristic of having been the result both the limited proteolysis of step 1. as well as of the complete fragmentation in the digestion step 3. for the determination of the conformational state.

[0010] In classical LiP-MS, native proteins are cleaved by an unspecific protease for a short period of time, which results in structure-specific peptide patterns. Mass spectrometry is used to identify these structural fingerprints. LiP-MS has been successful in capturing structural alterations on a proteome-wide scale in a plethora of applications. For example, it has enabled the identification of a new type of structural biomarkers of diseases, the global detection of targets of drugs and small molecule metabolites and the development of new analytical (“omic”) workflows for the detection of altered pathways. The standard LiP-MS workflow captures protein structural states in cell lysates.

[0011] Schopper. et al.: "Measuring protein structural changes on a proteomewide scale using limited proteolysis-coupled mass spectrometry", Nature Protocols, vol. 12, no. 11 , 26 October 2017, pages 2391-2410 reports on protein structural changes induced by external perturbations or internal cues and that these can profoundly influence protein activity and thus modulate cellular physiology. Limited proteolysis-coupled mass spectrometry (LiP-MS) is reported as an approach that enables the identification of protein structural changes directly in their complex biological context on a proteome-wide scale. After perturbations of interest, proteome extracts are subjected to a double-protease digestion step with a nonspecific protease applied under native conditions, followed by complete digestion with the sequence-specific protease trypsin under denaturing conditions. This sequential treatment generates structure-specific peptides amenable to bottom-up MS analysis. Next, a proteomics workflow involving shotgun or targeted MS and label-free quantification is applied to measure structure-dependent proteolytic patterns directly in the proteome extract. Possible applications of LiP-MS are reported to include discovery of perturbation- induced protein structural alterations, identification of drug targets, detection of disease- associated protein structural states, and analysis of protein aggregates directly in biological samples. The approach also enables identification of the specific protein regions involved in the structural transition or affected by the binding event. Sample preparation takes approximately 2 d, followed by one to several days of MS and data analysis time, depending on the number of samples analyzed.

[0012] Ma R. et al.: "Chemo-selection strategy for limited proteolysis experiments on the proteomic scale", Anal. Chem., vol. 90, no. 23, 7 November 2018, pages 14039-14047 describes a chemo-selective enrichment strategy, termed the semitryptic peptide enrichment strategy for proteolysis procedures (STEPP), to isolate the semitryptic peptides generated in mass spectrometry-based proteome-wide applications of limited proteolysis methods. The strategy involves reacting the e-amino groups of lysine side chains and any N-termini created in the limited proteolysis reaction with isobaric mass tags. A subsequent digestion of the sample with trypsin and the chemo-selective reaction of the newly exposed N-termini of the tryptic peptides with N-hydroxysuccinimide (NHS)-activated agarose resin removes the tryptic peptides from solution, leaving only the semitryptic peptides with one nontryptic cleavage site generated in the limited proteolysis reaction for subsequent LC-MS / MS analysis. As part of this work, the STEPP technique is interfaced with two different proteolysis methods, including the pulse proteolysis (PP) and limited proteolysis (LiP) methods. The STEPP-PP workflow is evaluated in two proof-of-principle experiments involving the proteins in a yeast cell lysate and two well-studied drugs, cyclosporin A and geldanamycin. The STEPP-LiP workflow is evaluated in a proof-of-principle experiment involving the proteins in two cell culture models of human breast cancer, MCF-7 and MCF- 10A cell lines. The STEPP protocol increased the number of semitryptic peptides detected in the LiP and PP experiments by 5- to 10-fold. The STEPP protocol not only increases the proteomic coverage, but also increases the amount of structural information that can be gleaned from limited proteolysis experiments. Moreover, the protocol also enables the quantitative determination of ligand binding affinities.

[0013] Heusel M. et al.: "Complex-centric proteome profiling by SEC-SWATHMS", Mol. Syst. Biol., vol. 15, no. 1 , Article no. e8438, 14 January 2019, pages 1-22 describe an integrated experimental and computational technique to quantify hundreds of protein complexes in a single operation. The method consists of size exclusion chromatography (SEC) to fractionate native protein complexes, SWATH / DIA mass spectrometry to precisely quantify the proteins in each SEC fraction, and the computational framework CCprofiler to detect and quantify protein complexes by error-controlled, complex-centric analysis using prior information from generic protein interaction maps. Our analysis of the HEK293 cell line proteome delineates 462 complexes composed of 2,127 protein subunits. The technique identifies novel sub-complexes and assembly intermediates of central regulatory complexes while assessing the quantitative subunit distribution across them. They make the toolset CCprofiler freely accessible and provide a web platform, SECexplorer, for custom exploration of the HEK293 proteome modularity.

[0014] WO-A-2023 / 099341 discloses a method for the detection of a conformational state of a protein, said protein being in a complex mixture of further proteins and other biomolecules, wherein the protein in said complex mixture has been subjected to a condition inducing a structural change in said the protein, comprising the following sequence of steps: 1. Limited proteolysis of the extract mixture under a condition in which the protein is in the original conformational state to be detected leading to a first fragment sample; directly followed by 2. Removal of large peptides and proteins or other biomolecules from said first fragment sample to form an enriched fragment sample; 3. Analytical analysis of the enriched fragment sample for the determination of fragments characteristic of having been the result of the limited proteolysis of step 1. as well as remaining after the removal step 2. for the determination of the conformational state of said at least one protein.

[0015] Van Kelly et al in Mol Cell Proteomics, 2022 Jan; 21(1); 100169. doi: 10.1016 / j.mcpro.2021.100169; Epub 2021 Nov 4 report, that comprehensive proteome analysis of rare cell phenotypes remains a significant challenge. A method for low cell number MS-based proteomics is proposed using protease digestion of mildly formaldehyde- fixed cells in cellulo, which they call the "in-cell digest." They combined this with averaged MS1 precursor library matching to quantitatively characterize proteomes from low cell numbers of human lymphoblasts. About 4500 proteins were detected from 2000 cells, and 2500 proteins were quantitated from 200 lymphoblasts. The ease of sample processing and high sensitivity makes this method suited for the proteomic analysis of rare cell states, including immune cell subsets and cell cycle subphases. To demonstrate the method, they characterized the proteome changes across 16 cell cycle states (CCSs) isolated from an asynchronous TK6 cells, avoiding synchronization. States included late mitotic cells present at extremely low frequency. They identified 119 pseudoperiodic proteins that vary across the cell cycle. Clustering of the pseudoperiodic proteins showed abundance patterns consistent with "waves" of protein degradation in late S, at the G2&M border, midmitosis, and at mitotic exit. These clusters were distinguished by significant differences in predicted nuclear localization and interaction with the anaphase-promoting complex / cyclosome. The dataset also identifies putative anaphase-promoting complex / cyclosome substrates in mitosis and the temporal order in which they are targeted for degradation. They demonstrate that a protein signature made of these 119 high-confidence cell cycle- regulated proteins can be used to perform unbiased classification of proteomes into CCSs. They applied this signature to 296 proteomes that encompass a range of quantitation methods, cell types, and experimental conditions. The analysis assigns a CCS for 49 proteomes, including correct classification for proteomes from synchronized cells.

[0016] Schopper et al in Nat Protoc, 2017 Nov; 12(11): 2391-2410; doi: 10.1038 / nprot.2017.100; Epub 2017 Oct 26, report that protein structural changes induced by external perturbations or internal cues can profoundly influence protein activity and thus modulate cellular physiology. A number of biophysical approaches are available to probe protein structural changes, but these are not applicable to a whole proteome in a biological extract. Limited proteolysis-coupled mass spectrometry (LiP-MS) is a proteomics approach that enables the identification of protein structural changes directly in their complex biological context on a proteome-wide scale. After perturbations of interest, proteome extracts are subjected to a double-protease digestion step with a nonspecific protease applied under native conditions, followed by complete digestion with the sequence-specific protease trypsin under denaturing conditions. This sequential treatment generates structure-specific peptides amenable to bottom-up MS analysis. Next, a proteomics workflow involving shotgun or targeted MS and label-free quantification is applied to measure structure-dependent proteolytic patterns directly in the proteome extract. Possible applications of LiP-MS include discovery of perturbation-induced protein structural alterations, identification of drug targets, detection of disease-associated protein structural states, and analysis of protein aggregates directly in biological samples. The approach also enables identification of the specific protein regions involved in the structural transition or affected by the binding event. Sample preparation takes approximately 2 d, followed by one to several days of MS and data analysis time, depending on the number of samples analyzed. Scientists with basic biochemistry training can implement the sample preparation steps. MS measurement and data analysis require a background in proteomics.

[0017] Bekei Beata in the NMR thesis "In-cell NMR Spectroscopy in Mammalian Cells" of 2013 reports, that while in vivo protein behaviours inside cells are influenced by many unique physical parameters, such as macromolecular crowding and intracellular viscosity, they are also affected by a plethora of biological activities that often vary greatly in different cell types. In-cell NMR spectroscopy is an invaluable biophysical tool for residue-specific, structural and functional investigations of isotope-labeled proteins inside native environments of cells and, as such, a preferred tool for studying intracellular proteins behaviours. The primary goal of the thesis was to extend the general applicability of eukaryotic in-cell NMR measurements to also include mammalian cells and to develop a generic protocol for the efficient delivery of isotope-labeled proteins into mammalian cells. Results demonstrate that low transduction efficiencies, high cell line dependences and vesicular-like intracellular distributions strongly limited the suitability of CPP-mediated protein delivery attempts. Although SLO- and EP-procedures yielded comparable transduction efficiencies at low applied protein concentrations, EP clearly outperformed the SLO approach at higher protein concentrations, because it enabled the linear delivery of increasing concentrations of exogenous proteins, with high correlations in intracellular cellular protein levels. In the second part of the thesis, a first-time description of the structural, dynamic and functional details of a human amyloid protein, alpha synuclein and its isoforms beta- and gamma synuclein inside different mammalian cells is given, as determined by high-resolution in-cell NMR experiments. Alpha synuclein, a small (14.5 kDa) intrinsically disordered neuronal protein is known to accumulate into cytosolic filamentous inclusions of dopaminergic neuronal cells in the course of neurodegenerative disorders termed synucleinopathies. High-resolution in-cell NMR studies were performed to provide novel insights into the native in vivo properties of this protein, especially in relation to its physiological and pathological properties, such as intracellular aggregation, membrane association and post-translation modifications. The choice of mammalian cells models included disease-unrelated (i.e. HeLa and A2780 cells) and disease-related dopaminergic neuronal cell lines (i.e. B65 and SK-N- SH cells). High-resolution 2D 1 H-15N SOFAST-HMQC spectra of the Synuclein protein family revealed monomeric appearances in five different mammalian cell lines, with no indications for significant conformational change, such as amphipatic helical conversions, or aggregations into beta- sheet rich, high molecular weight oligomers, or amyloid fibrils. The excellent spectral qualities of the Synuclein protein family in-cell NMR specimens obtained under healthy physiological conditions, are reported to hold great promises for incell NMR studies under pathogenic conditions related to Parkinson’s disease and the development of cell based in-cell NMR models for investigating the intracellular aggregation behaviours of these-, and other proteins.

[0018] SUMMARY OF THE INVENTION

[0019] Limited proteolysis-coupled mass spectrometry (LiP-MS) captures structural alterations to adapt to cellular perturbations on a proteome-wide scale. In LiP-MS, proteins in a native lysate are cleaved by an unspecific or low-specificity protease for a short period of time, which results in structure-specific peptide patterns that are analyzed by mass spectrometry. However, cell lysis may expose proteins to non-native conditions due to loss of subcellular compartmentalization and intracellular gradients, and dilution may disrupt labile interactions. This is for example detrimental for structures such as phase separated protein assemblies, that are held together by weak interactions.

[0020] Protein structures are highly dynamic and adapt to their environment and interaction partners. Techniques to determine high resolution structures of proteins typically require purification of proteins or protein complexes. However, perturbation-specific protein dynamics is only insufficiently captured using purified proteins. The cell provides a densely crowded environment in which proteins change with physical parameters like temperature or pH. In vitro settings cannot mimic complex intracellular molecule gradients. Protein structures may change according to their subcellular localization, and low affinity interaction partners are lost during cell lysis and purification. An example of this are stress granules that are labile macromolecular assemblies of proteins and RNA that form under stress conditions and are poorly characterized at the structural level. Therefore, it is important to study protein structural alterations in their native environment.

[0021] Cellular perturbations often involve cascades of multiple proteins in the time of seconds to minutes. These alterations involve the binding of small regulatory molecules, post- translational modifications, and changes in interaction partners. All these events affect protein structures, whereas changes in protein abundance occur on longer time scales. The complex interplay of protein structural alterations in a pathway largely impacts cell physiology.

[0022] To circumvent alteration of protein structures that may occur upon cell lysis and study protein structures in their cellular context, a novel method which we call in-cell LiP-MS was developed.

[0023] Until now it was not possible to apply LiP-MS to intact cells, so it was not possible to conduct the limited proteolysis reaction within cells, under physiological conditions that preserve molecular crowding, intracellular interactors, intracellular gradients, and compartmentalization. This is a serious limitation of the current LiP-MS workflow as cell lysis is likely to cause alterations of at least some protein structures and will certainly expose proteins to non-physiological conditions (e.g., dilution of the cellular content, loss of interactors, loss of subcellular context etc.). In-cell LiP-MS overcomes this limitation.

[0024] Other techniques currently enable the study of intracellular protein structures. Protein surface painting techniques, such as hydroxyl radical foot printing (HRF)-MS or demethylation of lysine residues detects changes in surface accessibility of proteins by oxidation with hydroxyl radicals. This method was applied to intact cells or organisms like C. elegans with residue level resolution. It employs a microfluidic system and laser setup, while in-cell LiP-MS uses a commercially available electroporation device. Further, the method currently shows lower reproducibility and substantially lower coverage compared to LiP-MS. Thermal proteome profiling (TPP) provides information on proteins that undergo changes in their intracellular stability, but, in contrast to in-cell LiP, it provides no structural details, i.e. the regions changing in a protein’s sequence / structure cannot be determined. In addition, to collect the soluble fraction, the cells are lysed before centrifugation, which leads to a loss of information under physiological conditions.

[0025] In-cell nuclear magnetic resonance (NMR) based methods and Forster resonance energy transfer (FRET) techniques provide information on the dynamics of protein structures inside cells, but typically one protein is studied at a time. In-cell LiP-MS provides peptide-level resolution, but for thousands of proteins simultaneously.

[0026] Compared to protein painting techniques, in-cell LiP-MS has substantially higher proteome and protein sequence coverage and currently higher reproducibility. Compared to TPP, it enables probing of structural features with peptide level resolution. Compared to standard LiP-MS it overcomes the need for cell lysis and consequent alterations of protein structures and loss of physiological conditions. Compared to FRET or NMR techniques, it enables probing of thousands of proteins simultaneously, as compared to probing only the structure of a single protein of interest and does not require tagging or labelling (although it does not enable determining the 3D coordinates of protein structures as NMR does).

[0027] To preserve the native environment and to study protein structures in their cellular context, this in-cell LiP-MS protocol is thus proposed herein. Using e.g. electroporation, a low- specificity protease, proteinase K, is introduced into cells.

[0028] It is shown that in-cell LiP-MS captures not only known specific structural changes like rapamycin-binding of FKBP1A, but also downstream effects of pathway activation. Moreover, in-cell LiP-MS captures the formation of phase separated protein assemblies, that are held together by weak interactions. Using in cell LiP-MS, we generated the first in situ dataset on the global structural alterations of proteins during stress granule formation upon sodium arsenite treatment. Structural changes of known stress granule components like G3BP1 are observed, and a resource was generated on peptide-level structural changes for hundreds of other proteins in response to sodium arsenite treatment. In addition, SERBP1 and PPP1 R12A are suggested to be involved in the early stages of cellular response to arsenite stress. Moreover, in-cell LiP provides evidence that the phase- separated compartment of nuclear speckles change upon sodium arsenite treatment, which we validate by fluorescence microscopy.

[0029] With in-cell LiP-MS a novel method is proposed to study a broad range of biological processes in their native environment, with peptide-level resolution, and on a proteome wide scale.

[0030] So for the purpose of studying protein structures in their native environment, a step to introduce a low-specificity protease, such as proteinase K, into living, for example by increasing the permeability of the cell membrane via electroporation, is used. This preferably happens under optimized conditions (see below). If incubated within cells for a short time, proteinase K cleaves proteins in solvent accessible and flexible regions. This results in protein fragments that report on a specific protein structural state. Cells are then lysed, the extracted fragments are optionally further processed with a protease like trypsin to make their size amenable to mass spectrometric analysis while retaining the structural information, and then identified and quantified using a proteomics workflow based on mass spectrometry. The novel combination of a step to increase the permeability of the cell membrane, e.g. electroporation, a method commonly used to study single proteins with incell nuclear magnetic resonance (NMR) spectroscopy, with established limited proteolysis coupled mass spectrometry (LiP-MS) allows to study thousands of protein structures simultaneously in their native context and how they adapt to various biological perturbations.

[0031] In a typical exemplary experiment, cells are treated with active compounds, exposed to stress conditions, or subjected to any other perturbation of interest. Cells growing on a cell culture dish are detached and resuspended in a physiological buffer (e.g., phosphate buffered saline, PBS). Next, a protease (e.g., a low specificity protease like proteinase K) is added to the cell suspension at a high concentration (e.g., 100 pM). Immediately afterwards, cells are electroporated to permeabilize the cell membrane. This allows the protease to diffuse into cells. After electroporation, the cells are incubated for a short time at a physiological temperature (usually 1-5 min at 37°C). The cells are then optionally washed by centrifugation with a physiological buffer to remove extracellular molecules. A lysis buffer containing a high concentration of chaotropes (e.g., 7 M guanidinium chloride) is then added and cells are heated to quench the protease reaction. The cells are lysed, fragments are optionally trypsinized to reduce their size and peptides are analyzed by mass spectrometry. Data from treated and control cells are quantitatively compared and fragments whose amounts differs between conditions are identified and mapped to protein sequences and (when available) to protein structures. The analysis reveals which proteins change their structure in response to the perturbation of interest and pinpoints the altered region.

[0032] The number of pulses and voltage of the electroporation can be optimized to allow maximal diffusion of the protease without impairing cell physiology and viability. The concentration of the protease and incubation time after electroporation can be optimized to obtain a maximum number of reproducible structure-specific cleavages. After electroporation, structure specific cleavages increase significantly in the presence of proteinase K on thousands of proteins, while cleavages occur mainly at the cell surface if the cells are not electroporated. The electroporation-dependent uptake of proteinase K in cells of interest can be confirmed using a fluorescent Forster resonance energy transfer (FRET) based protein sensor known to be cleaved by proteinase K. For example, the fluorescent proteins ECFP and YPet can be connected by a flexible peptide linker. Upon irradiation of ECFP, energy is internally transferred to YPet and YPet fluorescence is detected. When cells containing this sensor are incubated with proteinase K and electroporated, enabling proteinase K to diffuse into the cells, the fluorescent substrate is cleaved, and the cleavage is monitored by the loss of YPet fluorescence and increasing ECFP fluorescence. However, the substrate stays intact if proteinase K is added but not electroporated into the cells.

[0033] The novel in-cell LiP-MS method captures not only known structural changes of proteins, but provides additional biological information compared to established LiP-MS methods using cell extracts. For example, in HEK293 cells treated with 100 nM or 10 pM rapamycin of the drug rapamycin for 5 min or 10 min, in-cell LiP-MS detects the known structural change at FKBP1A, the protein target binding rapamycin. In addition, proteins involved in downstream biological processes also exhibit structural changes that are not observed if LiP-MS is applied to a native lysate. Hence, in-cell LiP-MS allows to capture biological events as they occur under physiological conditions. In the case of drugs, it allows capturing target engagement within the cell and pathways underlying the mode of action of the compound.

[0034] Moreover, in-cell LiP-MS enables probing the structural features and structural transitions of labile protein complexes, that are typically disrupted upon cell lysis. For example, in-cell LiP-MS was applied to study stress granule formation in mammalian cells. Stress granules are labile assemblies of proteins and RNA that form under stress conditions. HEK293 cells were treated with 500 pM sodium arsenite for 10 min, 20 min and 90 min and compared to untreated cells. With in-cell LiP-MS, structural changes of various known stress granule proteins were observed upon induction of stress granule formation, indicating that the approach detects the assembly of stress granules. Gene ontology enrichment analysis showed that structural changes were enriched in stress granule associated proteins after 20 min and 90 min of sodium arsenite treatment. At 10 min of arsenite treatment, when stress granules are not yet visible by fluorescence microscopy, in-cell LiP-MS already identified structural changes for a few specific stress granule proteins.

[0035] Essential to in-cell LiP-MS is the delivery of a protease into cells, coupled to quenching of protease activity and / or cell lysis, followed by the standard remaining components of a LiP- MS or Dark-LiP-MS workflow. In particular delivery via electroporation is shown. Possible variants of the protocol that enable conducting the limited proteolysis step within cells are the delivery of the protease by cell permeabilizing reagents such as detergents or toxins, coupling the protease with reagents with cell penetrating properties (e.g., peptides), delivery within caging systems and vesicles, intracellular expression of the protease in an active state under control of tight promoters, expression of an inactive version of the protease that can be activated when needed (e.g., photocaged protease, expression in the presence of an inhibitor or protease coupled to degron systems). Quenching of protease activity can be achieved by various conditions that denature proteins or inhibit its enzymatic activity, such as an increase in temperature, transfer of the cells into media that lyse cells and denature proteins simultaneously (e.g., mixture of mass spectrometry-compatible detergents and denaturants, highly concentrated acid solutions such as trichloroacetic acid), cell lysis in the presence of protease inhibitors. Variants of the remaining components of the LiP-MS workflow have been described in the previous patent describing the LiP-MS invention.

[0036] So far, in-cell LiP-MS has been tested for HEK293 cells, but the method also allows for studying any other cell type, including mammalian cell lines, primary cells, tissue, bacteria, yeast, and others. Settings are optimized for batch electroporation of 3 million cells, but electroporation settings can be adapted to allow down scaling, even to electroporation of single cells. Proteinase K is used because of its high activity at physiological temperature, but any other protease commonly used for structure specific proteolytic cleavage may be employed (e.g., subtilisin, thermolysin, papain). Optionally, a targeting sequence for specific subcellular locations can be coupled to the protease to be introduced into cells, to enable probing of the structural features of specific subcellular compartments (e.g., nuclear targeting sequencing).

[0037] LiP-MS allows to identify targets of drugs or other compounds within cells and under conditions of interest. The method simultaneously provides information downstream of the protein-target binding event and could be used to identify regulated pathways and off target effects. For example, the cellular response to a drug may be different in cells grown under hypoxic conditions occurring within the tumor environment. In-cell LiP could be used to detect target engagement, downstream and off-target events under these conditions. Moreover, in-cell LiP-MS can be used to determine optimal drug concentrations and may report on the time necessary for drug uptake and target engagement. It can also help to elucidate effects of mutations in pathways affected by drug binding. In addition, in-cell LiP- MS can be used to study how cells react if two or more drugs are added simultaneously or subsequently.

[0038] Besides drug binding, in-cell LiP-MS can also identify protein structural changes due to specific cellular conditions, for example changes in temperature or pH, which also occurs in the tumor microenvironment. LiP-MS may also provide information on global cellular response to other environmental factors, toxins, and irradiation.

[0039] In-cell LiP-MS can provide mechanistic information on any cellular process involving structural changes, including viral entry, the binding of modified T-cells in immunotherapy, or nutrient shifts.

[0040] Generally speaking, the proposed invention relates to a method for the determination of at least one protein or peptide or of properties thereof (preferably simultaneously for the structural and / or physico-chemical property determination more than one protein or peptide, such as more than 100 or more than 1000 proteins and / or peptides), said at least one protein or peptide contained in a complex mixture of further proteins and / or other biomolecules in at least one cell, comprising the following steps:

[0041] 1. delivery of a protease into said (preferably living) cell and limited proteolysis of the complex mixture, followed by cell extraction and / or cell lysis leading to a fragment sample (preferably this step is carried out by applying (only) an electrical field to said cells in order to increase the permeability of the cell membrane, in particular by using (pulsed) electroporation);

[0042] 2. denaturation of the fragment sample to a denaturated fragment sample;

[0043] 3. optional complete fragmentation of the denaturated fragment sample in a digestion step to a completely fragmented sample;

[0044] 4. analytical analysis of the fragmented sample or the completely fragmented sample for the determination of said at least one protein or peptide or of properties thereof.

[0045] The analytical analysis of the fragmented sample or the completely fragmented sample of this step 4 provides for an in-cell readout of the structural and / or physico-chemical properties of a protein or peptide. Preferably the analytical analysis is mass spectrometrybased.

[0046] The method is in particular for the determination of at least one structural and / or physicochemical property of said at least one protein or peptide.

[0047] As mentioned, preferably step 1. is carried out by applying an electrical field to cells in order to increase the permeability of the cell membrane, an electrostatic field is applied in a cell solution. Preferably (pulsed) electroporation is applied, for example applying at least one or a multitude of pulses e.g. pulses of more than 300 V / cm or more than 500 V / cm, e.g. each having preferably a length in the range of 20 - 100 ms (typically the voltage used is in the range of 500-1500V, normally in the range of 800-1200V).

[0048] Note that the denaturation of step 2. is not necessary and can also be omitted, then together with step 3. The denaturation of step 2. is in particular of interest in cases where one proceeds to the second digestion step 3. If one wishes to analyze the big fragments of step 1 directly by mass spectrometry one may skip denaturation and digestion, i.e. steps 2. and 3. completely.

[0049] So, the process may also be a method for the determination of at least one protein or peptide or of properties thereof (in particular for the determination of at least one structural and / or physico-chemical property of said at least one protein or peptide or the plurality thereof), said at least one protein or peptide contained in a complex mixture of further proteins and / or other biomolecules in at least one cell, comprising the following steps:

[0050] 1. delivery of a protease into said (preferably living) cell and limited proteolysis of the complex mixture, followed by cell extraction and / or cell lysis leading to a fragment sample (preferably this step is carried out by applying (only) an electrical field to said cells in order to increase the permeability of the cell membrane, in particular by using (pulsed) electroporation);

[0051] 4. analytical analysis of the fragment sample for the determination of said at least one protein or peptide or of properties thereof. Preferably the analytical analysis is mass spectrometry-based.

[0052] Notably, this approach is fundamentally different from classical proteomics approaches to measure protein abundances (not structures), as e.g. reported in the above paper of Van Kelly et al.: Since their approach is aimed at measuring protein abundances, native protein structures don’t need to be preserved. Therefore, they permeabilize cells with 90% methanol which partly dissolves cell membranes and affects protein structures. Therefore, this step precludes protein structural analyses. In addition, Kelly et al. do a benzonase treatment, which disrupts all interactions between proteins and DNA / RNA, thereby further affecting the possibility of analyzing native protein structures. Further the cells used in Kelly et al. are dead. They are formalin fixed and, in addition, permanently permeabilized. In the proposed approach here cells are preferably alive. After e.g. a short pulse electroporation, the pores on the cell membrane are closing up again and cells continue living. Importantly, Kelly et al. do classical trypsin digestion to completion (very long time, 16 hours). That means Kelly et al. again do not digest based on the structures of proteins, but based on the presence of specific amino acids in the sequence. Trypsin is a sequence-specific protease, proteinase K is not. Thus in 16 hours, trypsin will cut everywhere, roughly every 10 residues in a protein independent on the structure, as one does in classical proteomics analyses to measure abundances. In the proposed setup here, the time-limited digestion step e.g. with proteinase K (normally in the minute range) makes the cleavage structure-specific (proteinase K will cut only at the most structurally accessible regions of each protein). Further, Kelly et al. do only one digestion step with trypsin, again like in every classical proteomics analyses that measures protein abundances. The proposed workflow involves the double digestion e.g. with proteinase K followed e.g. by trypsin. The first digestion e.g. with proteinase K is based on the structure, the second one only reduces the size of the fragments to make them compatible with mass spec.

[0053] Also, the proposed approach is fundamentally different from NMR approaches only allowing for the structural tracking of one protein / peptide, as the proposed approach here allows and preferably is for structural screening of not just one but more than one, preferably tens, hundreds or thousands of peptides / proteins simultaneously and determining which of these peptides / protein(s) (if any) undergo structural alterations under the conditions of interest, or just determining their structural features.

[0054] As concerns the aspect, that the approach is particularly suitable for the determination of at least one structural and / or physico-chemical property of said at least one protein or peptide or the plurality thereof the following: structural properties also include conformational properties, which highlights the potential for aspects of structural dynamics. From the determination of such structural properties, one can often detect “functional” alterations of proteins. In other words, one can detect structural alterations via the proposed approach and from this information one can derive that the protein involved underwent also a functional alteration (e.g. altered activity if talking about enzymes). So the mentioned structural and / or physicochemical properties include corresponding associated functional properties of said at least one protein or peptide or the plurality thereof.

[0055] The proposed approach thus allows to determine / derive structural and / or biophysical properties, detects structural differences at specific protein regions, and thus detects proteins with altered functionalities. If one combines that resultant information with modelling tools such as AlphaFold, one can even derive 3D coordinates of atoms in the structure.

[0056] According to a first aspect of the invention it relates to a method for in-cell property determination of proteins or peptides by using an in-cell limited proteolysis by way of delivery of a protease into said cell. Using that approach structural and / or physical-chemical properties of a protein or peptide can be determined in situ in a cell, so under the natural biological environment. So, for the first time this proposed method allows to derive structural in-cell information directly using in particular mass spectrometric analytical analysis of fragmented samples. This can be useful for various purposes, e.g. assessing whether incell structures resemble in vitro structures obtained from purified proteins, or if proteins in a cell bind a drug in the same way as the corresponding purified protein.

[0057] However, the approach can also and in particular be applied if it is about comparing properties of proteins and / or peptides in cells as a function of or in reaction to an external or internal perturbation applied to the cell or a group of cells.

[0058] According to a second aspect of the present invention, it therefore relates to a method for the detection of a change of a state of a protein or peptide in response to a perturbation, wherein the said protein or peptide is contained in a complex mixture of further proteins and / or other biomolecules in at least two cells, comprising the following steps:

[0059] 1. subjecting at least one of said cells to a perturbation, followed or accompanied by delivery of a protease into said cell and by limited proteolysis of the complex mixture, followed by cell extraction and / or cell lysis leading to a fragment sample;

[0060] 2. denaturation of the first fragment sample to a denaturated fragment sample;

[0061] 3. complete fragmentation of the denaturated first fragment sample in a digestion step to a completely fragmented sample (this step is optional); carrying out steps 1.-3. with at least one other of said cells without subjecting that at least one other cell to said perturbation in step 1. for the generation of a completely fragmented control sample or a fragmented control sample,

[0062] 4. analytical analysis of the fragmented sample or the completely fragmented sample as well as of the completely fragmented control sample or fragmented control sample, respectively, for the determination of said change of state of the protein.

[0063] The terminology "change of state" of at least one protein or peptide is to be understood broadly as commonly accepted in the field, meaning a change of the sequence and / or the arrangement in space of the protein's constituent atoms determining the overall shape of the molecule or a change in the physico-chemical features of the protein or one of its regions (e.g. hydrophobicity or solvent accessibility changes due to binding of a small molecule to the protein). In other words, the expression "change of state" is not restricted to conformational changes but includes any kind of change information including changes in primary structure as well as going beyond the information of the primary structure, i.e. the linear sequence of amino acids and potential chemical modifications thereof in a peptide or protein. The expression "change of state" therefore includes the primary as well as secondary structure changes (three dimensional arrangement of local segments of proteins, the two most common secondary structural elements are alpha helices and beta sheets, but this also includes beta turns and omega loops), the supersecondary structure (motives, compact three-dimensional protein structure of several adjacent elements of a secondary structure that is smaller than a protein domain or a subunit), the tertiary structure (domains, three dimensional shape of the protein) and the quaternary structure (structure of proteins which are themselves composed of two or more smaller protein chains, also referred to as subunits). Drugs, when they bind to proteins, sometimes do not induce conformational changes in proteins. They only change properties of the protein such as solvent accessibility and this is sufficient to alter the degree of fragmentation induced by proteinase K. This is why using the proposed approach one can detect protein-drug interactions, one of the important applications of the proposed in cell LiP, so change of state also broadly includes such changes of properties.

[0064] Conformational changes in proteins, which can be identified using the proposed method, are made possible by their intrinsic flexibility. These changes may occur with only relatively small expenditure of energy. At the molecular structural level, conformational changes in single polypeptides are the result of changes in main chain torsional angles and side chain orientations. The overall effect of such changes may be localised with reorientations of a few residues and small torsional changes in the regional main chain. On the other hand torsional changes localised at very few critically placed residues may lead to large changes in tertiary structure. The later type of conformational changes is described as domain motions. The conformational changes associated with protein-protein interactions vary from local changes in side-chains rotameric states to global changes in structure such as collective domain movements.

[0065] When in step 1 mention is made of the perturbation being followed or accompanied by delivery of a protease into said cell, this includes situations where the perturbation of the cell happened earlier (for example patient biopsies or cells with mutations), and the limited proteolysis is only carried out significantly later. It also includes situations where the perturbation is applied somewhat before or also essentially immediately before the delivery of a protease, i.e. the perturbation is directly followed by the delivery of a protease into the cell. But it also includes situations where the perturbation is applied essentially at the same moment or concurrently with delivery of the protease into the cell and limited proteolysis is initiated. This is for example the case if the perturbation which is applied is actually applied at the same time and using the same mechanism as used for the delivery of a protease into set cell, so for example if electroporation is used for introducing the protease and at the same time also using electroporation the perturbation is introduced into the cell. It is thus also possible to use the timing between the perturbation and the start of the limited proteolysis to determine the time dependency of induced changes. This time dependency can be monitored as a function of the time elapsed between initiation of the perturbation and delivery of the protease into the cell and limited proteolysis for the situation where first the perturbation is applied and then the limited proteolysis started, or, in particular if the timescale of the effect of the perturbation is shorter than the timescale of limited proteolysis, it is possible to monitor the changes under limited proteolysis conditions by starting the limited proteolysis and initiating the perturbation in different experiments as a function of time after the start of the limited proteolysis. Note, that, as detailed below, if the protease is delivered into the cell in a inactivated form, which can be activated in situ in the cell by applying a corresponding stimulus, the wording also includes situations where delivery of the inactivated protease takes place well before applying the perturbation, and the perturbation is then applied followed or accompanied by limited proteolysis initiated by applying that stimulus activating the protease.

[0066] The expression cell in the present context includes mammalian cells, but also any other kind of animal cells, bacterial cells or plant cells, so it includes cells with or without cell walls, eukaryotic and prokaryotic cells. Also, the expression includes mature differentiated cells but also stem cells or progenitor cells. In particular the method is of interest if applied to differentiated mammalian cells, to mammalian progenitor cells, or mammalian stem cells. Cells can e.g. be fibroblasts, neurons, epithelial cells, hepatocytes, lymphocytes and macrophages.

[0067] The analysis in step 4. is normally based on a quantitative comparison of the analytical analysis of the completely fragmented sample or fragmented sample with the analytical analysis of the completely fragmented control sample or fragmented control sample, respectively.

[0068] For the analytical analysis in step 4., preferably mass spectrometry techniques, such as selected / multiple reaction monitoring (SRM / MRM), data-independent acquisition (DIA) of product ion spectra, including SWATH-MS and / or data-dependent acquisition (DDA or shotgun), is used.

[0069] A proteolytic system for step 1. can be selected from the group of (low-specificity) proteases such as protease K, Thermolysin, Subtilisin, Pepsin, Papain, a-Chymotrypsin, Elastase, Ficin, Pronase, and mixtures thereof. Note, that these proteolytic systems can be supplied and delivered into the cell in active form, but also in a form which has been inactivated. In the latter case limited proteolysis can be initiated in step 1 by applying a stimulus activating the corresponding protease in situ in the cell. Furthermore, the proteolytic systems can also be expressed inside the cell, so the expression delivery in the present context includes situations where the protease systems are not as a whole introduced into the cell but are rather expressed in the cell or formed in the cell using another mechanism (e.g. introducing fragments followed by assembly in the cell).

[0070] The proteases used are preferably low specificity proteases, so normally they are characterized in that they act as exopeptidases as well as endopeptidases and / or in that they cleave at more than two or more than three, or more than four or five different types of amino acids.

[0071] The delivery of the protease in step 1. can be implemented by a step of increasing the permeability of the cell membrane, preferably by using electroporation.

[0072] Step 1. can be directly followed by a step of removal of large peptides and proteins or other biomolecules from said fragment sample to form an enriched fragment sample. In said step of removal of large peptides and proteins or other biomolecules peptides and proteins or other biomolecules are preferably removed in a filtration, separation or another enrichment step, including size filtering; chromatography including size exclusion, hydrophobic or anion exchange chromatography; physical removal including phase separation, absorption, precipitation; filtration, separation or enrichment based on hydrophilic / hydrophobic properties; filtration, separation or enrichment based on electric / magnetic field; or a combination thereof.

[0073] In said step of removal of large peptides and proteins or other biomolecules peptides, proteins and / or other biomolecules having a molar weight larger than 20 kDa, preferably having a molar weight larger than 15 kDa, most preferably having a molar weight larger than 10 kDa can be removed from the first fragment sample.

[0074] Said perturbation is preferably selected from the group consisting of: temperature change; pressure change; ionic strength change; pH change; metabolic stimulant change; ligand addition, including drug / small molecule addition, metabolite addition, protein addition, peptide addition, lipid addition, DNA addition, RNA addition, vitamin addition, viral entry, internalization of molecules, irradiation, disease / health state or status and genetic variations, including mutations, or a combination thereof; addition of a chaotrope; induction of chemical modification, including post-translational modifications, in particular phosphorylations, disulphide bridge formation, ADP-ribosylation, ubiquitination, SUMOylation, acetylation, methylation, oxidation, glycosylation, or a combination thereof.

[0075] Step 4. typically includes, before actual analysis, a proteomics workflow, in particular involving denaturation, C18 cleanup, phase-separation, triggers or disruptors of interactions between molecules, triggers of protein degradation, or a combination thereof.

[0076] In step 1. the proteolytic system is preferably used at a concentration, with respect to the total biomolecular content in the sample, given as the ratio of enzyme to biomolecular content, in the range of 1 / 50-1 / 10000, preferably in the range of 1 / 100-1 / 1000 by weight.

[0077] In terms of molar concentration, the proteolytic system is preferably used at a concentration in the range of 5-1000 pM, preferably in the range of 20-500 pM, wherein this is the concentration in the LiP sample including the cells. For example, in case of proteinase K, a concentration of 100 pM is well suited, it works well in the range of 50 pM to 200 pM, even up to 500 pM. Reproducibility and coverage may suffer at higher concentrations under these conditions. These values are suitable for an electroporation volume of 100 pl of approx. 3 million HEK293 cells, electroporated with one 25 ms pulse at 1000 V. If higher voltage or more pulses are used, the optimal concentration may be lower. Other cell types may require different conditions.

[0078] Preferably the proteolytic system for step 1. is added to the cells (which may be provided in the form of a cell suspension or cell arrangement, including tissues and the like) in the form of an aqueous solution or suspension. Preferably the solution or suspension is added with a temperature below 30°C, preferably below 10°C, more preferably below 5°C. Typically the proteolytic system solution or suspension is stored on ice until edition. The temperature values are referred to the temperature of the proteolytic system solution or suspension and not to the temperature of the receiving cell suspension or arrangement.

[0079] Furthermore preferably, the proteolytic system is added and delivery, i.e. mechanisms to introduce or activate the proteolytic systems in(to) the cells, in particular electroporation, is initiated shortly after the proteolytic system addition. Typically, introduction of the proteolytic systems into the cells (for example by electroporation) is initiated or even terminated less than 20 seconds after addition of the proteolytic system, preferably not more than 10 seconds after addition of the proteolytic system.

[0080] Electroporation parameters may depend on cell type and local setup. To identify the best conditions for the system of choice, electroporation parameters can be adjusted aiming at reaching a sufficient amount of intracellular cleavages, while retaining cell viability and / or impacting cell physiology as little as possible, if viability and the physiological state of cells are important for the experiment in question. Cell viability can e.g. be assessed by staining cells after electroporation with trypan blue, a dye that cannot cross intact cell membranes of living cells. The percentage of living cells for some experiments may be controlled not to decrease after electroporation. A change in the trypan blue viability assay may also imply that cells leak. Moreover, electroporation can lead to changes in cell morphology, that may indicate physiological alterations. For example, under electroporation at too high intensity / voltages it may be observed under the light microscope that cells clump (3 pulses of 1400 V). Lastly, changes in the physiology of cells induced by electroporation can also be detected by mass spectrometry or mass spectrometry-based proteomics. This also means the that also the electroporation step or differences in the electroporation may be used as the above-mentioned perturbation. On the other hand, to exclude that electroporation introduces significant changes (e.g. pathway alterations, or cleavages within protein sequences), electroporated cells and control cells without electroporation can be analyzed by mass spectrometry after protein extraction and trypsinization and peptide profiles quantitatively compared to detect changes in peptide or protein levels.

[0081] Step 1. is preferably carried out for the limited proteolysis phase over a time span of 1-60 minutes, preferably in the range of 2-30 minutes, or 2 - 10 minutes or 2 - 5 minutes, further preferably at a temperature in the range of 20-40°C.

[0082] In step 1. the limited proteolysis phase can preferably be terminated by quenching of protease activity, in particular by inducing conditions that denature proteins or inhibit its enzymatic activity, including an increase in temperature, transfer of the cells into media that lyse cells and denature proteins simultaneously.

[0083] Also in step 1. the limited proteolysis phase can be followed by a washing step, preferably using a saline solution.

[0084] The at least one protein is preferably a protein based exclusively on proteinogenic amino acids, or is based on proteinogenic amino acids and carries post-translational modifications. Further, the invention relates to the use of such a method for the determination, in a hypothesis-free manner, of a change of state, in particular of a conformation or an alteration of physico-chemical properties, of said at least one protein, said at least one protein having undergone a change of state, in particular conformational changes or alterations of physicochemical properties, after perturbation induced in the investigated complex mixture, or of a medically relevant state, in particular conformation, of the protein, for the determination of protein-based drugs, for the influence of drugs or other ligands on proteins, or for quality control of protein-based pharmaceutical preparations.

[0085] Preferably this can take place in combination with peptide fragment enrichment techniques such as TAILS (terminal amine isotopic labeling of substrates) for the peptides generated by step 1.

[0086] Further embodiments of the invention are laid down in the dependent claims.

[0087] BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Preferred embodiments of the invention are described in the following with reference to the drawings, which are for the purpose of illustrating the present preferred embodiments of the invention and not for the purpose of limiting the same. In the drawings,

[0089] Fig. 1 shows the In-cell LiP-MS workflow, (a) The low-specificity protease proteinase K is introduced into cells by electroporation. By incubating the cells for a short time, structure-specific peptides are generated. After protease inactivation, the peptides are analyzed by mass spectrometry, (b) Electroporation of 200 pM proteinase K leads to significantly increased half-tryptic peptides (bright) and decreased fully tryptic peptides (dark) (5 replicates, adj. p-value < 0.01 , 2-fold changes), (c) Addition of 200 pM proteinase K without electroporation does not lead to significantly increased half-tryptic peptides (bright) and decreased fully tryptic peptides (dark) (5 replicates, adj. p-value < 0.01 , 2-fold changes), (d) Electroporation without addition of proteinase K does not lead to significantly increased half-tryptic peptides (bright) and decreased fully tryptic peptides (dark) (5 replicates, adj. p-value < 0.01 , 2-fold changes), (e) Proteinase K- induced cleavages indicated that significantly increasing half-tryptic peptides and decreasing fully-tryptic peptides occur mainly on cytosolic and nuclear proteins, (f) The intensity of half-tryptic peptides increases with added proteinase K and is higher after electroporation, (g) Cleavages per protein at the optimized settings of 100 pM added proteinase K, 1 pulse of 25 ms at 1000 V electroporation, followed by 2 min limited proteolysis at 37°C. In total, 4263 distinct PK cleavage sites on 1656 proteins were detected (adj. p-value < 0.01 , 2-fold changes).

[0090] Fig. 2 shows the validation of proteinase K delivery and target detection, (a) Electroporation of proteinase K into HEK293 cells expressing an ECFP / YPET FRET sensor of protease activity leads to intracellular cleavage. Addition of proteinase K without electroporation, and the process of electroporation itself are not sufficient, (b) Structural changes in cells treated with 100 nM rapamycin dissolved in DMSO compared to cells treated only with DMSO (6 replicates). After 5 min of treatment, cells were electroporated with 100 pM proteinase K. Peptides of the rapamycin target FKBP1A are highlighted in middle shading, (c) Structure of rapamycin (bright) and the interacting protein FKBP1A (PDB 2DG3). Regions in dark are detected by mass spectrometry, region in middle shading is significantly changing upon treatment with rapamycin. (d) Structural changes in native lysates treated with 10 nM rapamycin dissolved in DMSO compared to cells treated only with DMSO (6 replicates). After 5 min of treatment, cells were proteinase K was added in a ratio of 1 :100 to total protein concentration. Peptides of the rapamycin target FKBP1A are highlighted in middle shading.

[0091] Fig. 3 shows an analysis of structural changes upon arsenite treatment of HEK cells, (a) HEK293 cells were treated with sodium arsenite, (b) HEK293 cells were stained with Anti-G3BP1 antibody and Hoechst staining after 90 min treatment with sodium arsenite, (c) Peptide level changes of HEK cells treated with sodium arsenite compared to untreated cells (6 replicates, q-value < 0.05, 1.5-fold changes), (d) Gene ontology enrichment analysis of proteins with significantly changing peptides.

[0092] Fig. 4 shows structural changes upon sodium arsenite treatment of HEK293 cells, (a) Structural changes on selected proteins upon arsenite treatment. The barcode plots depict the protein sequence, middle shaded regions are detected by mass spectrometry. Light shaded regions change significantly at the indicated time (6 replicates, q-value < 0.05, 1.5-fold changes), (b) Alphafold structure of G3BP1. Middle shaded regions are detected by mass spectrometry. Light shaded regions change significantly at the indicated time (6 replicates, q-value < 0.05, 1.5-fold changes).

[0093] Fig. 5 shows how nuclear speckles become more round upon sodium arsenite treatment of HEK293 cells, (a) HEK293 cells were stained with Anti-SC35 antibody and Hoechst staining after 90 min treatment with sodium arsenite, (b) Circularity of nuclear speckles with and without sodium arsenite treatment was measured for three biological replicates. Corresponds to immunostained images from A. The colours correspond to replicates;

[0094] Fig. 6 shows target detection and reproducibility of in-cell LiP-MS; (a) HEK293 cells treated with 20 pM rapamycin dissolved in DMSO for 10 min compared to intensities in cells treated only with DMSO in an in-cell LiP setup. Left, middle and right plots show three independent experiments. Peptide intensities, each data point represents a single peptide; peptides of FKBP1A are shown in black. The lines marks significance levels (FC > 1.5, p-value < 0.01 , n = 6 replicates), (b) Peptide intensities in a native lysate of HEK293 cells treated with 10 nM rapamycin dissolved in DMSO for 5 min compared to intensities in cells treated only with DMSO (6 replicates). Each data point represents a single peptide; peptides of FKBP1A are shown in red. The shaded gray region marks significance levels (FC > 1 .5, p-value < 0.01 , n = 4 replicates). The four FKBP1 A peptides with the lowest p-value are highlighted, (c) Number of peptides with missing quantities per treatment in (a,b). (d) Median coefficient of variation (CV) of peptide quantities per treatment in (a,b). (e) Intensity of half-tryptic peptides relative to total peptide intensity for in-cell LiP samples of rapamycin treated HEK293 cells, (f) Overall sequence coverage considering peptides detected in at least 3 replicates per condition;

[0095] Fig. 7 shows a comparison of methods detecting changes upon glucose starvation of mammalian cells, (a) Metabolite intensities from HEK293 cells treated with glucose-free medium for 2 h compared to untreated cells. Each data point represents a single metabolite. The shaded gray region marks significance levels (log2(FC) > 0.5, adjusted p-value < 0.05, n = 3 replicates), (b, c, d) Peptide intensities from HEK293 cells treated with glucose-free medium for 2 h compared to untreated cells using the indicated method. Each data point represents a single peptide. The shaded gray region marks significance levels (FC > 1.5, q-value < 0.05, n = 6 replicates). Mitochondrial peptides are highlighted in black, (b) Standard LiP-MS in native lysates, (c) Peptide quantities in (b) corrected for changes in protein abundances, (d) In-cell LiP-MS of intact cells (e, f) Gene ontology enrichment analysis (p-value < 0.01 ; Biological process) of proteins with significantly changing peptides in (c, d);

[0096] Fig. 8 shows structural changes upon 2 h glucose starvation, (a) Standard LiP-MS: Phosphoglycerate mutase dimer (PDB: 2JKV). Dark grey regions are covered by mass spectrometry analysis, bright grey regions (dotted arrows) are significantly changing (FC > 1.5, q-value < 0.01 , n = 6 replicates). The second subunit is highlighted in light grey, phosphoglycerate binding sites are highlighted in dark grey, (b) In-cell LiP-MS: Pyruvate kinase tetramer (PDB: 4GPZ). Black regions are covered by mass spectrometry analysis, bright grey regions (dotted arrows) are significantly changing (FC > 1.5, q-value < 0.01 , n = 6 replicates). The second subunit is highlighted in white, fructose-1 ,6- bisphosphate binding sites are highlighted in mid grey, (c) In-cell LiP-MS: Fructose-bisphosphate aldolase (PDB: 5KY6). (d) In-cell LiP-MS: Phosphogluconate dehydrogenase dimer (PDB: 2JKV). Black regions are covered by mass spectrometry analysis, bright grey regions (dotted arrows) are significantly changing (FC > 1.5, q-value < 0.01 , n = 6 replicates). The second subunit and regions not covered by mass spectrometry analysis are highlighted in light grey, NADP+ binding sites are highlighted in mid grey;

[0097] Fig. 9 shows changes upon glucose starvation, (a) Gene ontology enrichment analysis (adjusted p-value < 0.01) of metabolic pathways with significant changes of metabolites upon 2 h glucose starvation, (b) Gene ontology enrichment analysis (p-value < 0.01 ; Biological process) of proteins with significantly changing peptides upon glucose starvation using standard LiP-MS. (c-h) Comparison of HEK293 cells treated with glucose-free medium for 2 h compared to untreated cells and analzyed with the indicated method, (c) Each data point represents a single peptide. Mitochondria-associated proteins are highlighted in grey. The shaded gray region marks significance levels (FC > 1.5, q-value < 0.05, n = 6 replicates), (d) Overall sequence coverage considering peptides detected in at least 3 replicates per condition, (e) Intensity of half-tryptic peptides relative to total peptide intensity, (f) Median coefficient of variation (CV) of peptide quantities per treatment, (g) Number of peptides with missing quantities per treatment, (g) Each data point represents a single protein. The shaded gray region marks significance levels (FC > 1.5, q-value < 0.05, n = 6 replicates).

[0098] DESCRIPTION OF PREFERRED EMBODIMENTS

[0099] To expand the high— reproducibility and specificity global protein structural analyses of LiP- MS to living cells, we are delivering the low-specificity protease proteinase K into mammalian cells. Various protein delivery techniques have been applied for in-cell NMR and Cas9 delivery for gene editing, including cell-penetrating peptides linked covalently to proteins, pore-forming toxins, osmocytosis, extracellular vesicles, and electroporation. Comparison of delivery methods in In-cell NMR showed that for in-cell NMR applications, electroporation is more efficient than pore-forming toxins and cell penetrating peptides. Extracellular vesicles and osmocytosis operate on a time scale of hours, while electroporation happens within milliseconds. This makes electroporation a suitable method for LiP-MS;, in which exact timing matters. Electroporation permeabilizes cellular membranes for a brief period and therefore also allows for reproducible protein uptake efficiency than the other methods with less well characterized routes of protein internalization.

[0100] Electroporation has previously been used for delivery of aSyn for in-cell NMR, proteins of the kinetochore for fluorescence study of fuctional proteins and for delivery of ere recombinase for genome editiding across Arabidopsis cell wall.

[0101] We show that by combining LiP-MS with electroporation, we can study protein structural changes in living cells on a proteome-wide scale.

[0102] Validation: FRET and Rapamycin

[0103] Using a Forster resonance energy transfer (FRET) based sensor, we confirm that active proteinase K is delivered into cells by electroporation. We validate that in-cell LiP-MS captures rapamycin binding specifically to the protein FKBP1A. This interaction was originally reported for yeast (Heitman 2001), and it is well detected by standard LiP-MS in native lysates of human cells (Piazza 2020). We show that this interaction is also captured by in-cell LiP-MS.

[0104] Novel application: stress granules

[0105] Finally, we apply in-cell LiP-MS to study the formation of stress granules. Stress granules are dynamic assemblies of proteins and RNA, that form under stress conditions such as sodium arsenite treatment, oxidative stress or heat shock to prevent mRNA degradation. Using in-cell LiP-MS, we obtained a dataset on the global structural alterations at different time points during sodium arsenite treatment. We detect a known structural transition on the stress granule core protein SGBP1. In addition, in cell LiP-MS suggests that SERBP1 and PPP1R12A are involved in the early stages of cellular response to arsenite stress. Moreover, we provide evidence that nuclear speckles become rounder in shape during sodium arsenite treatment.

[0106] Results

[0107] Method development Workflow: To study protein structural states in their native environment, we tested whether electroporation introduces proteinase K into mammalian cells. HEK293 cells were detached from a culture dish, washed and resuspended in PBS (Fig. 1a). Upon addition of proteinase K (PK), cells were immediately permeabilized by electroporation. Cells were subsequently incubated for 1 min at a physiological temperature of 37°C for structure specific cleavage of intracellular proteins. Extracellular proteinase K was removed by washing with PBS, and proteinase K activity was quenched by addition of a chaotrope (7M guanidinium chloride) and heating to 98°C for 5 min. The samples were snap-frozen. Cells were thawed on ice and lysed by sonication. After reduction and alkylation of disulfide bonds, trypsin was added to the lysate, which was measured by quantitative, label-free MS analysis.

[0108] PK cleavages increase after electroporation: We used proteolytic cleavage as readout of PK delivery into cells. In contrast to trypsin which cleaves preferably after lysine and arginine residues, PK has low sequence specificity. Hence, a peptide with one or two termini that does not match the trypsin recognition motif is more likely to be generated by PK cleavage. An increase in half-tryptic peptides, and a decrease in the fully tryptic counter peptides therefore reports on PK activity. We compared proteolytic cleavage in cells that were electroporated with PK to control cells, to which PK was added but not electroporated (Fig. 1b). Upon electroporation, half tryptic peptides increased, and fully tryptic peptides decreased, corresponding to PK cleavages on 1257 proteins. This suggests that PK indeed enters the cells after electroporation. This is supported by gene ontology enrichment analysis: proteins cleaved by PK showed a significant enrichment of cytosolic and nucleus- associated terms (Fig. 1e), indicating that proteinase K diffuses into cells upon electroporation. In contrast, addition of PK to cells without electroporation did not lead to increased proteolytic cleavage compared to cells without PK (Fig. 1c). Hence, addition of PK to cells without electroporation is not sufficient to increase proteolytic cleavage significantly.

[0109] Minimizing electroporation effects on cell physiology and viability: Next, we asked whether the process of electroporation introduces peptide-level changes. We compared electroporated HEK293 cells to cells non-electroporated control cells. No significant electroporation-induced changes were observed at one 25 ms pulse at 1000 V (Fig. 1 d), and at 1-3 pulses of 25 ms at 800 V. Fully-tryptic and semi-tryptic peptides were symmetrically distributed in all tested electroporation settings, suggesting that electroporation does not cause protein cleavage. At high voltage and pulse number, the observed peptide-level changes corresponded to significantly decreasing protein abundances, suggesting that high voltage induces protein aggregation. We assessed the effect of electroporation on cell viability by staining cells with trypan blue within 5 min after electroporation. After electroporation for 25 ms with 1 pulse of 1000 V, cell viability did not decrease significantly (ANOVA p adj. < 0.05. Electroporation of PK with one pulse of 25 ms at 1000 V was chosen for subsequent experiments to exclude that observed changes were due to electroporation.

[0110] No cell leakage after electroporation: Since pore formation by electroporation allows PK to diffuse into the cells, we tested whether intracellular proteins also diffuse into the supernatant. We measured the protein content of the cell supernatant after electroporation by BCA. Protein concentration in the supernatant did not increase after electroporation, suggesting that cell leakage is negligible.

[0111] Optimizing proteome coverage: In order to maximize proteome coverage, we investigated how PK concentration affects the number of intracellular cleavages. Extracellular PK concentrations of 100 pM lead to the highest number of reproducible intracellular cleavages after electroporation at one 25 ms pulse at 1000 V. The intensity of cleaved peptides increased at higher PK concentrations (Fig. 1f). At 1000 pM PK, half-tryptic peptides comprised up to 17% of total peptide intensity (compared to 30% half-tryptic peptides in standard LiP-MS experiments). However, reproducibility of PK cleavages decreased compared to 100 pM PK, possibly because of over-digestion. Hence, 100 uM PK was electroporated in subsequent experiments. At the optimized conditions, 4263 distinct PK cleavage sites on 1656 proteins were detected (Fig. 1g).

[0112] Increasing incubation time after electroporation from 1 min to 2 min, and six instead of five replicates further increased the number of reproducible of electroporation-dependent PK cleavages. Using six replicates per condition, 8045 peptides on 2463 of 4760 detected protein groups changed significantly, corresponding to 4263 distinct PK cleavage sites on 1656 proteins. Prolonging limited proteolysis to 5 min did not further increase PK cleavage. To conclude, electroporating 3 million HEK cells with 100 pM PK followed by incubation for 2 min at 37°C prior to washing lead to the highest number of reproducible PK-induced cleavages, covering over 1600 mainly intracellular proteins. These settings were used for all subsequent experiments. Because we did not impute missing values, half-tryptic peptides that are only generated intracellularly were not considered in this analysis. The actual coverage in an experiment comparing two perturbed cell states may be even higher. Method validation

[0113] Validation of intracellular delivery using FRET: We used HEK cells expressing a Forster resonance energy transfer (FRET) based ECFP-YPet sensor to confirm that proteinase K diffuses into cells during electroporation. Because of internal energy transfer from CFP to YPet, yellow YPet fluorescence indicates an intact sensor, and an increase of YPet fluorescence relative to ECFP fluorescence indicates cleavage at the linker region. Upon addition of PK to cells, the sensor stayed intact, which shows that PK does not enter the cells. After electroporation, the cyan fluorescence of the cleavage product increased, confirming that active PK enters the cells (Fig. 2a). Electroporation without addition of proteinase K did not lead to cleavage of the sensor.

[0114] Valdiation of target detection: Rapamycin binding: To validate that limited proteolysis in cells captures known binding events, we treated HEK293 cells with rapamycin, which binds specifically to the protein FKBP1A. After treatment with 100 nM rapamycin for 5 min, cells were electroporated with proteinase K and processed following the previously optimized workflow. A peptide mapping to FKBP1A was found among the most significant peptides (6 replicates; Fig. 2b, mapping close to the known rapamycin binding site (Fig. 2c). A similar result was obtained for cells treated with 10 pM rapamycin for 10 min prior to electroporation of proteinase K. We found 61 peptides on 60 proteins changing significantly (6 replicates, adj. p-value < 0.05, 1.5-fold changes). One of these peptides mapped to FKBP1A. Hence, limited proteolysis in cells captures a known binding event in a highly complex native background.

[0115] Next, we compared LiP in cells to a standard LiP-MS experiment. A native lysate of HEK203 cells was treated with 10 pM rapamycin for 5 min prior to addition of PK. Multiple peptides of FKBP1A were found among the most significant peptides (4 replicates; Fig. 2d). In LiP in cells, the percentage of half-tryptic peptides was lower than in the native lysate, indicating less proteinase K cleavage. Proteinase K was among the most abundant proteins in both methods. Variation was similar in both approaches, even though replicates of in-cell LiP- MS were electroporated and processed one at a time, and all lysate samples were treated at the same time.

[0116] Absolute quantification of proteinase K: Next, we determined the number of proteinase K molecules entering the cell upon electroporation. Three million HEK293 cells were mixed with a final concentration of 100 pM proteinase K in 120 pl, electroporated with one 25 ms pulse at 1000 V and incubated at 37°C for 2 min. Samples were then washed twice with PBS and proteinase K was inactivated. The same workflow was repeated without electroporation for control cells. Before mass spectrometry analysis, proteinase K specific peptides labeled with heavy amino acids were added to the samples. By comparing to a calibration curve, and taking into account that only a fraction of the samples was digested, we calculated that the electroporated sample contained 52.5 pmol proteinase K, and the control sample 16.0 pmol proteinase K. We assumed the difference of 36.6 pmol proteinase Kto be intracellular (of 12 mmol added to the cells). With three million cells, this corresponds to 12.2 amol proteinase K per cell, or 7.3 million molecules. For HeLa cells, the number of proteins been estimated at 2.0 * 10A9 molecules per cell. If we assume the number of proteins in HEK cells to be similar, there are approximate 274 substrate proteins per intracellular proteinase K molecule. With a cellular volume of approximately 2000 pm3, this corresponds to 6.1 pM intracellular proteinase K and 1.7 mM substrate proteins.

[0117] Additional application: stress granules

[0118] HEK293 cells are treated with sodium arsenite: After we observed that In-cell LiP-MS captures the high affinity interaction of rapamycin and FKBP1A in situ, we asked whether it also detects structural changes during the formation of low affinity stress granules. To test this, we treated HEK cells with sodium arsenite, which induces the formation of stress granules. Cellis were analyzed in six replicates 0 min, 10 min, 20 min and 90 min after addition of arsenite. Cells detached from the culture dish in PBS pre-warmed to 37°C, centrifuged at 200xg for 30 s, resuspended in PBS and immediately electroporated with proteinase K. The samples were then processed according to the previously optimized incell LiP protocol (Fig. 3a). Immunofluorescence staining of G3PB1 confirmed that stress granules form between 10 min and 20 min of arsenite treatment (Fig. 3b).

[0119] In-cell LiP-MS captures structural changes during stress granule formation: After 10 min of treatment, before the appearance of stress granules by microscopy, we detected structural changes on three peptides, mapping to three proteins (q-value < 0.05, 1.5-fold change, Fig. 3c). At 20 min of arsenite treatment, 2325 peptides on 1149 proteins changed significantly. At 90 min, 20841 peptides 3146 proteins changed significantly. With few exceptions, protein levels did not change significantly, and structural changes were therefore not adjusted for protein abundance changes.

[0120] Although PK activity was higher at later time points, gene ontology enrichment of proteins changing structure after 20 min and 90 min of arsenite treatment showed, that proteins exhibiting structural changes are indeed enriched in stress granules (Fig. 3d). Moreover, RNA binding and RNA related biological processes are significantly enriched. This confirmed that in-cell LiP-MS reliably identifies labile macromolecular assemblies that undergo structural changes upon perturbation with sodium arsenite.

[0121] In-cell LiP captures known structural change on G3BP1 with peptide-level resolution: Next it was asked, whether in-cell LiP provides reliable information on protein and even peptide level. Therefore, we looked at the structural changes detected for known stress granule associated proteins. Many showed structural changes after 20 min and 90 min of arsenite treatment, among them G3BP1 , a core component of stress granule assembly (Fig. 4a).

[0122] Stress granule formation is triggered by a structural change in G3BP1 in response to free RNA. G3BP1 consists of a NTF2-like domain required for dimerization, an RNA recognition motif, and three intrinsically disordered regions (IDRs). Both dimerization and RNA binding are essential for stress granule assembly. It has been suggested that G3BP1 is usually present in a closed conformation, held together by electrostatic interactions between IDR1 and IDR3. When free RNA levels increase during arsenite stress, RNA competes for binding on the RNA recognition motif and IDR3. RNA-bound G3BP1 switches to an open state leading to liquid-liquid phase separation and providing the core of stress granules.

[0123] Interestingly, we detect a structural change on the RNA recognition motif after 20 min of arsenite treatment, which is in line with the proposed model of G3BP1. At 90 min, structural changes occur on all domains of G3BP1 (Fig. 4b).

[0124] In-cell LiP newly identifies SERBP1 and PPP1 R12A as they exhibit early structural changes: In fluorescence microscopy, stress granules appeared after 20 min of arsenite treatment. Interestingly, significantly changing LiP peptides on three proteins, SERBP1 , PPP1 R12A, and CKB were already detected after 10 min. To our knowledge, CKB has not been associated with stress granules before. PPP1 R12A has been found to associate with stress granules. SERBP1 has been reported to act as inhibitor of ribosomes upon mTOR phosphorylation. Upon arsenite treatment, SERBP1 localizes to both stress granules and nucleoli. A previous study reported that SERBP1 is required for clearance of stress granules after arsenite stress by modulating the degradation of G3BP1. A knockout of SERBP1 did not alter stress granules after 60 min of arsenite treatment, but changes may occur earlier. In-cell LiP suggests changes on nuclear speckles upon arsenite treatment

[0125] Interestingly, gene ontology enrichment analysis suggested that nuclear speckles and the spliceosome, which are located in nuclear speckles change significantly at 20 min of arsenite treatment. To investigate whether sodium arsenite indeed affects nuclear speckles, HEK293 cells were stained with Anti-SC35 antibody and Hoechst staining after 90 min treatment with sodium arsenite (Fig. 5a). We observed that nuclear speckles became rounder upon sodium arsenite treatment than in control cells (Fig. 5b). This suggests that in-cell LiP enables to identify novel macromolecular rearrangements in the cell.

[0126] Discussion

[0127] Comparison to other methods: Compared to established LiP-MS in native lysates, organelles, and intracellular molecule concentrations are preserved by using electroporation to introduce the low-specificity protease proteinase K into cells. It eliminates the need for lysis, and allows to study cellular physiology with minimal invasion. In-cell LiP- MS is a proteome-wide screening method to provide a basis for subsequent functional studies.

[0128] In comparison, HRF-MS reports on intracellular structural changes with residue-level resolution but requires a more complicated microfluidic setup. TPP, on the other hand, requires a simple temperature gradient to identify proteins that undergo a structural change, but is lacking information on the region of the change. In comparison to in-cell NMR which also uses electroporation, our method does not require labeling and is not limited to one or a few proteins at a time. However, in-cell LiP reports on structure dynamics with peptide level resolution, not atomic level, and can suffer more from false positives.

[0129] Workflow: Electroporation settings for in-cell LiP were optimized to preserve cell viability which allows us to study cellular physiology with minimal invasion. Proteinase K concentration and incubation time were chosen to maximize the number of reproducible cleavages. With the current settings, we cover several thousands of proteins, however, sequence coverage remains low on the majority of proteins. We show that coverage isn’t dramatically increased by changing proteinase K concentration and incubation time. However, in applications where cell viability can be neglected, higher voltage and pulse number during electroporation may lead to more uniform influx of proteinase K and possibly to better sequence coverage by increasing reproducibility. However, proteinase K concentration and incubation time may have to be adjusted, and possible loss of cellular material taken into consideration.

[0130] In-cell LiP allows to study treatment of detached cells in solution, as shown with rapamycin, as well as adherent cells detached after treatment like in the stress granule study. We used HEK293 cells, but other mammalian cells may work as well, although viability and proteome coverage may have to be checked. In-cell LiP may even be applicable to tissues. Using stronger electroporation to permeabilize the cell wall, or pretreatment with wall degrading enzymes, it may also be possible to introduce a low specificity protease into bacteria or yeast.

[0131] If cellular material is limited, less cells may be used, the limit being the mass spectrometry detection limit. We show that cell leakage is no major issue after electroporation. Hence, washing may be reduced or even eliminated completely in future experiments. In-cell LiP- MS was optimized for a physiological temperature of 37°C, but it may be adapted to other temperatures, pH, or different low-specificity proteases. It is even feasible to introduce an inactivated protease by electroporation, and to start limited proteolysis at a later time.

[0132] Validation of intracellular delivery: We validated intracellular delivery of proteinase K using a fluorescent FRET based sensor expressed in HEK cells. After electroporation, proteinase K cleaved the linker region. The fluorophores appear to be resistant to cleavage by proteinase K, possibly because of their stable beta barrel fold. In future, this assay could be used to further quantify proteinase K activity in cells, and to check that cellular or incubation conditions don’t affect proteinase K activity. Moreover, the fluorescent assay can be used to efficiently screen electroporation conditions and proteinase K concentration on other cell lines. Further, we validated the approach by the detection of the interaction between the drug rapamycin and its known protein target.

[0133] In-cell LiP-MS conditions have been optimized to preserve the native state of cells as far as possible. However, after incubation with extracellular proteinase K, cells tend to clump, e.g., during centrifugation. Clumping during limited proteolysis is reduced, if the added proteinase K is ice-cold and electroporation is performed within seconds after mixing proteinase K and cells.

[0134] We determined that the intracellular concentration of proteinase K after electroporation is about 6.1 pM, which corresponds to one proteinase K molecule per approximately 274 substrate molecules. This is a rough estimation based on one replicate, and assuming that HEK cells are similar in size and protein number to HeLa cells, for which numbers were taken from literature. Moreover, this value represents the average of all electroporated cells. It would be interesting to determine the distribution of intracellular proteinase K concentrations after electroporation in single cells. However, flow cytometry is difficult to apply because of the abovementioned clumping after incubation with proteinase K. It is interesting to note that the estimated intracellular concentration of proteinase K of 6.1 pM is much lower than the extracellular concentration of 100 pM, which confirms that the cells stay largely intact and that electroporation-induced pores close.

[0135] In classical LiP-MS on lysates, optimal coverage has been found for 1 pg proteinase K added to 100 pg substrate protein. Because of the crowded environment inside the cell, a direct comparison of the ratio of proteinase K to substrate is of limited use. Inside the cell, certain regions of proteins, which are accessible in a native lysate, may be protected from proteolytic cleavage. Gene ontology enrichment suggested that cleavages occur mainly on cytosolic and nuclear proteins. In future applications it will be interesting to determine whether other organelles such as mitochondria or lysosomes are also covered by electroporation of proteinase K. Rapamycin.

[0136] Material and Methods

[0137] Cell culture conditions

[0138] HEK293 cells (ATCC #CRL-1573) were cultured in Dulbecco’s modified Eagle’s medium (DM EM, Gibco Life Technologies #41965039) supplemented with 10% heat inactivated fetal bovine serum and 5% Penicillin / Streptomycin at 37°C and 5% CO2 in a humidified environment. The HEK293 cells were passaged prior to confluency in phosphate-buffered saline (PBS) pH 7.4 (Gibco #10010015). To measure cell viablility and concentration, cells were mixed 1 :1 with trypan blue solution and analyzed in a Countess Cell Counter (Thermo Fisher Scientific).

[0139] Cell transfection

[0140] HEK239 cells were transfected at approx. 70% confluency on 150 mm culture dishes. Medium was exchanged prior to transfection. Per plate, 1200 pl jetPrime buffer were mixed with 25 ug plasmid DNA (ECFP-TevS-YPET, a gift from Charlie Morgan and Jason Chin; Addgene #100097), and 50 pl jetPrime reagent were added (Polyplus #101000046). After incubation for 10 min, the mixture was added to the cells.

[0141] Fluorescence-based proteinase K activity assay

[0142] HEK293 cells were detached from culture dishes 2-3 days after transfection, washed with PBS, and resuspended in PBS at a concentration of 33.3*106cells / ml. Prior to fluorescence analysis, 90 pl cell suspension (corresponding to 106cells) were mixed with 30 pl water or 30 pl of a 400 pM proteinase K solution PK (solubilized in water). For electroporation, 100 pl of the solution were electroporated with one 25 ms pulse at 1000 V using a Neon Transfection System (Thermo Fisher Scientific), and transferred to a 96-well plate. Control cells without electroporation were directly transferred after mixing proteinase K and cells. Fluorescence of ECFP (430-10 / 480) and FRET to YPet (430-10 / 530-10) were measured using a CLARIOstar Plus plate reader (BMG Labtech).

[0143] Rapamycin treatment of cells

[0144] HEK293 cells were detached from culture dishes, washed with PBS and resuspended in PBS at a concentration of 40*106cells / ml. For treatment, 75 pl cell suspension per replicate (corresponding to 106cells) were mixed with 15 pl rapamycin solution to a final concentration of 100 nM or 10 pM rapamycin (rapamycin dissolved in DMSO and PBS; 1 % DMSO concentration on cells in the 10 pM rapamycin experiment, and 0.016% in the 100 nM rapamycin experiment). After incubation for 5 min (for 100 nM rapamycin) or 10 min (for 10 pM rapamycin), cells were processed according to the in-cell LiP-MS workflow.

[0145] Sodium arsenite treatment of cells

[0146] HEK293 cells were grown in separate dishes for each replicate. Sodium arsenite dissolved in water was added to the medium in a final concentration of 500 pM. After 10 min, 20 min, and 90 min, the medium was removed, cells washed in PBS, and pelleted by centrifugation at 200xg for 1 min. Cells were resuspended in PBS. Part of the sample was snap-frozen in liquid nitrogen for analysis of protein abundances. For in-cell LiP, 90 pl containing approx. 106cells were processed according to the in-cell LiP-MS workflow.

[0147] In-cell LiP-MS workflow

[0148] For in-cell LiP-MS, 90 pl of treated or untreated cells (approx. 106cells) were mixed with 30 pl ice-cold proteinase K solution (100 pM final proteinase K concentration except in optimization experiments, proteinase K dissolved in water). To determine the effect of proteinase K on cells, 30 pl water were added instead. To determine the effect of electroporation independent of proteinase K, PBS was added instead. For electroporation, 100 pl of the cell suspension were electroporated using Neon Transfection System (Thermo Fisher Scientific). Electroporation was usually carried out with one 25 ms pulse at 1000 V, except for optimization experiments when 1-3 pulses and 800 V to 1400 V were tested. Exactly 15 seconds after electroporation, samples were heated to 37°C, and incubated for 2 min (during optimization, 1 min to 5 min incubation were tested). For washing, 1 ml PBS pre-warmed to 37°C was added, and cells were pelleted by centrifugation for 1 min at 4000xg. The supernatant was removed, and cells were washed a second time with 1 ml PBS. For protease inactivation, 7M guanidinium hydrochloride dissolved in water was added to the cells, and the samples were heated to 98°C for at least 5 min. Samples were snap-frozen in liquid nitrogen before cell lysis.

[0149] Cell lysis

[0150] In-cell LiP-MS: samples were thawed on ice, and sonicated two times 30 sec using a Hielscher UP200St-G - Ultrasonic generator coupled to a UP200St with VialTweeter (optimization experiments and 10 pM rapamycin experiment: 60 % cycle, 100 % amplitude; stress granule experiment, 100 nM rapamycin experiment; 170 W power, 60% cycle). Protein concentration was measured using a bicinchoninic acid assay and evenly adjusted. Samples were then subjected to tryptic digestion.

[0151] Measuring protein abundances

[0152] To measure protein abundances after arsenite treatment, cell pellets were thawed on ice, and 8 M urea buffer supplemented with protease inhibitor cocktail (Roche) was added. Cells were sonicated like the in-cell LiP-MS samples of the stress granule experiment, protein concentration was measured using a bicinchoninic acid assay and evenly adjusted. Samples were then subjected to tryptic digestion.

[0153] Classical LiP-MS: HEK293 cells were detached from culture dishes, washed with PBS and resuspended in a buffer consisting of 1 mM MgCI2, 150 mM KCI, 100 mM HEPES pH 7.4. Cells were lysed using a tissue grinder (DKW Life Sciences Kimble Pellet Pestle) in ten cycles of 10 s of homogenization and 1 min pause on ice. The lysate was cleared by centrifugation at 1000 x g at 4°C for 15 min. The supernatant was diluted to a protein concentration of 2 pg / ul and treated with rapamycin.

[0154] Classical LiP-MS workflow of rapamycin treatment

[0155] For classical LiP-MS, 50 ul native lysate of 2 pg / ul protein were incubated with 1 ul rapamycin (dissolved in DMSO), or 1 ul DMSO for 5 min at 25°C. To start limited proteolysis, 5 ul proteinase K (0.2 pg / ul in water) were added for 5 min. For protease inactivation, samples were heated to 99°C for 5 min, then cooled to 4°C for 5 min. Finally, an equal volume of a 10% sodium deoxycholate solution was added. Samples were subjected to tryptic digestion. Tryptic digestion

[0156] Disulfide bonds were reduced by incubation with 5 mM tris(2-carboxyethyl)phosphine hydrochloride) (TCEP) shaking for 40 min at 37 °C. Subsequently, samples incubated 30 min in the dark at room temperature with 40 mM. Samples were then diluted using 0.1 M ammonium bicarbonate (Sigma Aldrich) to achieve a final guanidinium chloride concentration of 0.5 M, or 1% sodium deoxycholate, or <2 M urea. Proteins were digested with Lys-C (FUJIFILM Wako Pure Chemical Corporation) and trypsin (Promega) at an enzyme substrate ratio of 1 :100 shaking overnight at 37 °C. The tryptic digest was stopped by adding formic acid to a final pH of approx. 2.

[0157] C18-Cleanup and MS sample preparation

[0158] A 96-well C18 Micro-Spin plate (The Nest Group, 40-400 pg) was washed with 200 pl methanol, followed by 100 pl buffer B (50 % acetonitrile (ACN), 0.1 % FA), and 2x 200 pl buffer A (0.1 % FA). Samples were loaded on the plate and washed with 3x 200 pl buffer A by centrifugation at 1000xg for 1 min. Peptides were eluted with 3x 100 pl buffer B and dried in a vacuum concentrator. Samples were resuspended in buffer A supplemented with iRT peptides (Biognosys) and analyzed by mass spectrometry.

[0159] Absolute quantification of proteinase K

[0160] Four synthetic peptides (AQUA Ultimate Heavy, lysines and arginines stable isotopelabeled, Thermo Scientific) that are proteotypic for proteinase K were selected for absolute quantification (peptides 1-4). Heavy labeled peptides were added at a final concentration of 500 fmol / pl before mass spectrometry analysis to an in-cell LiP sample electroporated previously with 100 pM proteinase K. Peptides were also added to a control sample to which proteinase K was added but not electroporated. For calibration, heavy peptides were spiked into a lysate containing no proteinase K (peptide concentrations 500 fmol / pl, 50 fmol / p, 5 fmol / pl, 500 amol / pl). Samples were measured by PRM. Peptide intensity was determined by summing up the area of the five most intense fragment ions. A linear model was fitted separately for each peptide to Iog10-transformed peptide intensities and concentrations of the calibration samples. The linear model was used to quantify the corresponding light proteinase K peptides in the samples of interest. Finally, proteinase K quantity was calculated as the mean of the four determined peptide quantities.

[0161] Liquid Chromatography and mass spectrometry data acquisition

[0162] DIA: Samples were injected into a Orbitrap Fusion Lumos Tribrid mass spectrometer equipped with a Waters nanoAQUITY UPLC System using a linear gradient from 3-35% B (Eluent A: 0.1 % formic acid, Eluent B: 99.9% acetonitrile, 0.1 % formic acid) over 120 minutes at a flowrate of 300 nL / min. The MS1 scans ranged from 350 to 1400 m / z with an Orbitrap resolution of 120000 and an AGO target of 50 % or 100 ms injection time. The DIA method consisted of 41 variable width windows with 1 m / z overlap. Fragments were generated by high energy collision induced dissociation (HCD) using a fixed collision energy of 28 %. MS2-DIA spectra were acquired over a scan range of 200 to 1800 m / z with an Orbitrap resolution of 30000 and 200 % AGO target or 54 ms injection time.

[0163] DIA: Samples for optimization of electroporation settings, PK concentration and time were injected into a Orbitrap Eclipse Tribrid mass spectrometer equipped with a Waters nanoAQUITY LIPLC System using a linear gradient from 3-30% B (Eluent A: 0.1 % formic acid, Eluent B: 95% acetonitrile, 0.1% formic acid) over 120 minutes at a flowrate of 300 nL I min. The MS1 scans ranged from 350 to 1400 m / z with an Orbitrap resolution of 120,000 and an AGO target of 800,000. The DIA method consisted of 41 variable width windows with 1 m / z overlap. Fragments were generated by high energy collision induced dissociation (HCD. MS2-DIA spectra were acquired over a scan range of 150 to 2000 m / z with an Orbitrap resolution of 30,000 and 50,000 AGC target.

[0164] DIA: The in-cell LiP-MS samples of 10 pM rapamycin were injected into a Q Exactive Plus Hybrid Quadrupole-Orbitrap mass spectrometer equipped with an EASY-nLC 1000 system (Thermo Fisher Scientific). Separation of peptides was conducted on a 40 cm x 0.75 mm i.d. column (New Objective, PF360-75-10-N-5) which was in house packed with 3 pm beads (Dr. Maisch Reprosil-Pur 120). Separation followed a linear gradient of 3%-30% buffer B over a total run time of 120 minutes, followed by 5 min with an isocratic constant concentration of 90% buffer B (Eluent A: 0.1 % formic acid, Eluent B: 95% acetonitrile, 0.1 % formic acid). The flow rate was kept at 300 nl / min. The MS1 scans ranged from 350 to 1500 m / z with an Orbitrap resolution of 70,000 and an AGC target of 3e6 or 120 ms injection time. The DIA method consisted of 20 variable width windows with 1 m / z overlap. Fragments were generated by high energy collision induced dissociation (HCD. MS2-DIA spectra were acquired over a scan range of 200 to 1800 m / z with an Orbitrap resolution of 35,000 and 1e6 AGC target.

[0165] PRM: Samples were injected into a Orbitrap Fusion Lumos Tribrid mass spectrometer equipped with a Waters nanoAQUITY UPLC System using a linear gradient from 3-35% B (Eluent A: 0.1 % formic acid, Eluent B: 99.9% acetonitrile, 0.1 % formic acid) over 120 minutes at a flowrate of 300 nL / min. The isolation window was 1.2 m / z. Fragments were generated by high energy collision induced dissociation (HCD) using a fixed collision energy of 28 %. MS2 spectra were acquired over a scan range of 200 to 1800 m / z with an Orbitrap resolution of 60000 and 1000 % AGC target or 118 ms injection time.

[0166] Search engine

[0167] The data was searched using Spectronaut 17 (Biognosys), with libraries built from the DIA files. The digestion type was set to semi-specific and peptide minimum length was adjusted to 6. The data was searched against the human reviewed UniProt database and the sequence of proteinase K (htps: / / www.uniprot.org, retrieved on April 16, 2023). Data analysis was performed with Spectronaut via a classical DIA analysis using a targeted library-based search with the following modifications: a single-hit was defined by modified sequence, and single-hit proteins were excluded. Quant 2.0 was selected for label-free protein quantification, samples were normalized globally on the median, and PTM localization was selected.

[0168] Statistical analysis

[0169] Only peptides with a quantity <1 were considered for statistical analysis using R version 4.2.2. A Welch’s t-test was used for statistical analysis. P-values were adjusted to multiple testing using the Benjamini Hochberg method.

[0170] Gene ontology enrichment analysis

[0171] For gene ontology enrichment analysis, the topGP package (version 2.50.0) was used in R (htps: / / git.bioconductor.org / packages / topGO). Gene ontology annotations were obtained from Uniprot (htps: / / www.uniprot.org, retrieved on May 17, 2022). Enrichment analysis was performed with the Fishers exact test using the elim algorithm in topGO. Significance was set at the p-value level < 0.01 .

[0172] Peptide mapping on A I ph a Fol d2 structures

[0173] Peptides were mapped onto structures predicted by Alphafold v4 using PyMOL (version 2.5.4).

[0174] Immunostaining and circularity measurements

[0175] HEK293 cells were seeded on 8 well p-Slides with ibiTreat surface (ibidi #80806) at 25000 cells per well. After 24 h 500 pM sodium arsenite (Pfaltz & Bauer #7784-46-5) was added to the cells. After 90 mins cells were fixed for 10 min with 4% PFA (Electron Microscopy sciences #15714) in PBS. After washing twice with PBS the cells were permeabilized for 10 min with PBS 0.5% Triton X-100 (Sigma-Aldrich #T8787). Cells were washed twice with PBS and then incubated simultaneously with primary antibody, secondary antibody and 1 :1000 Hoechst 33258 (Molecular Probes #H-3569) for 1 h in PBS with 0.1% Tween 20 (Sigma-Aldrich #P1379). Cells were washed afterwards 2 times in PBS with 0.1 % Tween 20 and kept in PBS for imaging.

[0176] To determine the circularity of nuclear speckles, a representative field of view of each replicate was taken and thresholded in imaged to create a mask of the nuclear speckles that matches the observed speckles by eye. Circularity was then measured via the analyse particles function.

[0177] Antibodies used:

[0178] As further proof of concept, the following further series of experiments was carried out:

[0179] In-cell LIP-MS reproducibly identifies structural changes upon drug binding

[0180] In-cell LiP-MS with delivery of a protease into said cell and limited proteolysis can capture known structural alterations associated to high-affinity binding events by making use of the well-characterized binding of rapamycin to the protein FKBP1 A (FKBP prolyl isomerase 1A, Uniprot P62942). After treatment with 20 pM rapamycin for 10 minutes, cells were electroporated with PK and processed following an optimized workflow. To assess the reproducibility of in-cell LiP-MS, we repeated the experiment two more times. In addition, we performed standard LiP-MS of a rapamycin-treated HEK293 cell lysate, to allow a comparison to the established method.

[0181] Using in-cell LiP-MS, we detected multiple FKPBIA-derived peptides among the strongest hits in a comparison of peptide intensities between rapamycin-treated and control (DMSO- treated) cells (Fig. 6a; FC > 1.5, p-value < 0.01 , n = 6 replicates). Identical FKBP1A peptides changed in two (Fragmentl , Fragment2) or in all three independent experiments (Fragments, Fragment4). Standard LiP-MS also identified multiple FKBPIA-derived peptides among the most significant hits (FC > 1.5, p-value < 0.01 , n = 4 replicates; Fig. 6b), as previously shown. The FKBP1A peptides detected by in-cell LiP-MS overlap with those detected by standard LiP-MS of treated lysate, but are longer. This is likely the result tryptic cleavage being more efficient in 1% sodium deoxycholate, which is used in standard LiP-MS, compared to 0.5 M guanidinium hydrochloride used in in-cell LiP-MS. Hence, incell LiP-MS identified identical peptides of the expected target protein among the top hits in three independent experiments.

[0182] Most peptides were quantified consistently across all replicates per condition (Fig. 6c). The median coefficient of variation was at 14-24% for peptides in in-cell LiP-MS, and 15-17% in standard LiP-MS (Fig. 6d). The percentage of HT peptides was close to 18% in all experiments (Fig. 6e). In this comparison, we used a similar ratio of protein substrate and PK in both conditions: i.e. , 100 pg protein and 35 pmol PK (1 pg) in the lysate and 3 million cells (800-1000 pg protein) and 12 nmol (347 pg) PK for in-cell LiP. Mass spectrometric sequence coverage of proteins was higher for in-cell LiP-MS experiments than the standard LiP-MS experiment, which results from the fact that different mass spectrometers were used to acquire the datasets (Fig. 6f; considering only peptides detected in at least three replicates per condition). We also performed power calculations on the first in-cell LiP-MS experiment and the standard LiP-MS experiment (In-cell LiP-MS: power = 34% at FC = 1.5 and power = 99% at FC = 2.0, with n= 6, FDR = 5%; standard LiP-MS: power = 46% at FC = 1.5 and power = 99% at FC = 2.0 with n = 4, FDR = 5%). Hence, data quality measures of in-cell LiP-MS are overall similar to those of the established in-lysate LiP-MS method.

[0183] In-cell LiP-MS identifies structural changes in glucose starvation

[0184] In-cell LiP-MS also detects other types of structural changes like oligomerization. To test this, HEK293 cells were incubated in glucose-free medium. After 2 h of glucose starvation, metabolomics measurements identified 21 metabolites that were changing significantly (Fig. 7a; log2(FC) > 0.5, adjusted p-value < 0.05, n = 3 replicates). Gluconeogenesis, glycolysis, and the pentose phosphate pathway were among the most significantly enriched pathways compared to untreated control cells (Fig. 9a). We performed in-cell LiP-MS under the same conditions, and measured protein abundances using a standard proteomics workflow. In parallel, cells were harvested after 2 h glucose starvation, disrupted by native lysis, and analyzed by standard LiP-MS. We also generated tryptic control samples, i.e. native lysate treated with water instead of PK to measure changes in protein abundance during standard LiP-MS.

[0185] Standard LiP-MS identified 375 changing peptides on 244 proteins, mainly associated with mitochondria (Fig. 9 b,c). In-cell LiP-MS, on the other hand, identified only three changing peptides (Fig. 9c; FC > 1.5, q-value < 0.05, n = 6 replicates), one peptide on Glycogenin-1 (GYG1), and two overlapping peptides on the enzyme pyruvate kinase (PKM). CVs, missing values, share of HT peptides and sequence coverage were similar or better than in the previous rapamycin study (Fig. 9d-g; power = 66% at FC = 1.5 and power = 99% at FC = 2.0, with n= 6, FDR = 5%). Since changes in PKM observed by in-cell LiP-MS were in accordance with the metabolomics measurements, which found its substrate phosphoenolpyruvate to be changing, we wondered about the hundreds of hits identified by standard LiP-MS. Intriguingly, the tryptic control of the standard LiP-MS experiment suggested a systematic downregulation of mitochondrial proteins after 2 h glucose starvation, which was not observed in abundance measurements performed with the standard proteomics workflow (Fig. 9d). Our previous analysis of organelle coverage showed that mitochondrial coverage is lower for standard LiP-MS than for in-cell LiP-MS, suggesting that mitochondria are insufficiently disrupted by native lysis. Human mitochondria have been reported to fuse into larger networks during glucose starvation. The native lysates were cleared by centrifugation at 10,000xg before standard LiP-MS, at which step mitochondria were likely removed to a larger extent from the starvation samples than from control samples due to their difference in size. We concluded that the mitochondrial- specific changes observed in standard LiP-MS upon glucose starvation are likely artefacts of native cell lysis that report on differential centrifugation behavior, but not necessarily on intracellular abundance changes of proteins.

[0186] To identify actual structural changes of proteins, we corrected peptide quantities in standard LiP-MS for protein-level changes in the tryptic control. The multiple testing correction of p- values was performed simultaneously on the standard LiP-MS data, the abundance corrected standard LiP-MS data and the in-cell LiP-MS data to obtain identical q-value levels, which facilitates the comparison (Fig. 7b-d). In-cell LiP-MS identified 11 significantly changing peptides on 7 proteins (3 peptides on Pyruvate kinase (PKM, Uniprot P14618), and on Glycogenin-1 (GYG1 , Uniprot P46976), and abundance-adjusted standard LiP-MS 22 peptides on 19 proteins (2 peptides on Uridine-cytidine kinase 2 (UCK2, Uniprot Q9BZX2), 3 peptides on the Elongation factor 1-alpha 1 (EEF1A1 Uniprot P68104). A GO enrichment analysis identified glycolysis as the most significant biological process in both methods (Fig. 7e,f). In-cell LiP-MS identified also the pentose phosphate pathway, which is in accordance with the metabolomics measurements. The only enzyme for which changes were detected by both in-cell LiP-MS and standard LiP-MS is Glycogenin-1 (GYG1), which initiates glycogen synthesis. In-cell LiP-MS also identified a structural change on the Long- chain-fatty-acid-CoA ligase 4 (ACSL4), which is involved in both lipid synthesis and degradation via beta-oxidation.

[0187] Among glycolytic enzymes, standard LiP-MS detected a structural change on one glycolytic enzyme, phosphoglycerate mutase-1 (PGAM, Uniprot P18669; Fig. 8a), a homodimer catalyzing the reversible conversion of 3-phosphoqlycerate to 2-phosphoqlycerate. Phosphoglycerate was also changing in the metabolomics dataset. The structural change occurs at the substrate binding region, suggesting that it reports on substrate occupancy. A different peptide located at the substrate binding region is changing in the in-cell LiP-MS dataset, but just above the significance cutoff (q-value = 0.056).

[0188] In-cell LiP-MS detected three changing peptides on pyruvate kinase (PKM), which catalyzes the final irreversible and rate-limiting step of glycolysis. PKM changes between an active tetrameric form, and inactive monomeric or dimeric states upon binding of the allosteric regulator fructose-1 ,6 bisphosphate. The changing peptides are located at the interface of the two dimers, in direct proximity to the fructose-1 ,6 bisphosphate binding sites (Fig. 7b), hence they report on enzyme regulation via oligomerization.

[0189] Another structural change was detected by in-cell LiP-MS on aldolase (ALDOA, Uniprot P04075). The changing peptide was neither located at the oligomer interface, nor close to the substrate binding site, which is to be expected for an enzyme that is active in both glycolysis and gluconeogenesis (Fig. 7c). Instead, the changing peptide was located at the surface of the multimer, which may indicate that interactions with other molecules change during glucose starvation. For example, it has been shown that in absence of fructose-1 ,6- bisphosphate, aldolase activates AMPK (AMP-activated protein kinase), thereby regulating protein translation during. Interestingly, the Eukaryotic translation initiation factor 4E-bindi ng protein 1 (EIF4EBP1 , Uniprot Q13541), a repressor of translation initiation, and the Small ribosomal subunit protein uS8 (RPS15A, Uniprot P62244), a component of the ribosome, were also found to be changing by in-cell LiP-MS.

[0190] Among the enzymes of the pentose phosphate pathway, in-cell LiP-MS identified a structural change in gluconate dehydrogenase (PGD), the main cellular enzyme for NADPH production. PGD has an inactive monomeric and active oligomeric form, which is regulated by the binding of NADP+. The peptide changing in in-cell LiP-MS after 2 h glucose starvation is linking the two homo-dimers, and adjacent to the regulatory NADP+ binding site (Fig. 8a). We concluded that while standard LiP-MS successfully captured substrate binding in PGM, which is expected to be a high-affinity interaction, in-cell LiP-MS may be more sensitive to oligomerization and complex formation, as shown in PKM, PGD, and potentially ALDOA.

[0191] Methods for second series of experiments in as far different from the above Methods In-cell LiP-MS workflow

[0192] For classical LiP-MS of glucose-starved cells, the 50 pl native lysate of 2 pg / pl protein were incubated for 5 min at 25 °C. To start limited proteolysis, 5 pl PK (0.2 pg I pl in water) were added for 5 min. For protease inactivation, samples were heated to 99 °C for 5 min, then cooled to 4 °C for 5 min. Finally, an equal volume of a 10% sodium deoxycholate solution was added. Samples were subjected to tryptic digestion.

[0193] Classical LiP-MS workflow

[0194] For classical LiP-MS of glucose-starved cells, the 50 pl native lysate of 2 pg / pl protein were incubated for 5 min at 25 °C. To start limited proteolysis, 5 pl PK (0.2 pg I pl in water) were added for 5 min. For protease inactivation, samples were heated to 99 °C for 5 min, then cooled to 4 °C for 5 min. Finally, an equal volume of a 10% sodium deoxycholate solution was added. Samples were subjected to tryptic digestion.

[0195] Liquid Chromatography and mass spectrometry data acquisition DIA: The in-cell LiP-MS samples of glucose starvation and 20 pM were injected into a Q Exactive Plus Hybrid Quadrupole-Orbitrap mass spectrometer and analyzed like described in the above methods for the 10 pM rapamycin dataset.

[0196] Statistical analysis In the classical LiP-MS experiment of glucose starvation, peptides were adjusted for changes in protein abundance using the MSstatsLiP package (version 1.9.2).

Claims

CLAIMS1. Method for the determination of at least one structural and / or physicochemical property of at least one protein or peptide, said at least one protein or peptide contained in a complex mixture of further proteins and / or other biomolecules in at least one cell, comprising the following steps:

1. delivery of a protease into said cell and limited proteolysis of the complex mixture, followed by cell extraction and / or cell lysis leading to a fragment sample;2. denaturation of the fragment sample to a denaturated fragment sample;3. optional complete fragmentation of the denaturated fragment sample in a digestion step to a completely fragmented sample;4. analytical analysis of the fragmented sample or the completely fragmented sample for the determination of said at least one protein or peptide or of properties thereof.

2. Method according to claim 1 for the detection of a change of a state of said protein or peptide contained in said complex mixture of further proteins and / or other biomolecules in at least two cells in response to a perturbation, comprising the following steps:

1. subjecting at least one of said cells to said perturbation, followed or accompanied by delivery of a protease into said cell and by limited proteolysis of the complex mixture, followed by cell extraction and / or cell lysis leading to a first fragment sample;2. denaturation of the first fragment sample to a denaturated fragment sample;3. optional complete fragmentation of the denaturated first fragment sample in a digestion step to a completely fragmented sample; carrying out steps 1.-3. with at least one other of said cells without subjecting that at least one other cell to said perturbation in step 1. for the generation of a completely fragmented control sample or a fragmented control sample,4. analytical analysis of the fragmented sample or the completely fragmented sample as well as of the completely fragmented control sample or fragmented control sample, respectively, for the determination of said change of state of the protein.

3. Method according to claim 2, wherein the analysis in step 4. is based on aquantitative comparison of the analytical analysis of the completely fragmented sample or fragmented sample with the analytical analysis of the completely fragmented control sample or fragmented control sample, respectively.

4. Method according to any of the preceding claims, where for the analytical analysis in step 4. mass spectrometry techniques, in particular selected / multiple reaction monitoring (SRM / MRM) and / or data-independent acquisition (DI A) of product ion spectra, including SWATH-MS and / or optionally data-dependent acquisition (DDA or shotgun), is used.

5. Method according to any of the preceding claims, wherein in the step 1., a proteolytic system selected from the group consisting of protease K, Thermolysin, Subtilisin, Pepsin, Papain, a-Chymotrypsin, Elastase, ficin, pronase, and mixtures thereof is used and / or wherein the proteolytic system for step 1 . is added to the at least one cell in the form of a preferably aqueous solution or suspension, wherein preferably the solution or suspension is added having a temperature below 30°C, preferably below 10°C, more preferably below 5°C.

6. Method according to any of the preceding claims, wherein the delivery of the protease in step 1. is implemented by a step of increasing the permeability of the cell membrane, preferably by using electroporation or by using permeabilizing reagents, or by coupling the protease with reagents with cell penetrating properties, by delivery within caging systems and / or vesicles, by intracellular expression of the protease in an active state under control of tight promoters, expression of an inactive version of the protease that can be activated when needed, or a combination thereof.

7. Method according to any of the preceding claims, wherein step 1. is directly followed by a step of removal of large peptides and proteins or other biomolecules from said fragment sample to form an enriched fragment sample, wherein in said step of removal of large peptides and proteins or other biomolecules, preferably peptides and proteins or other biomolecules are removed in a filtration, separation or another enrichment step, including size filtering; chromatography including size exclusion, hydrophobic or anion exchange chromatography; physical removal including phase separation, absorption, precipitation; filtration, separation or enrichment based on hydrophilic / hydrophobic properties; filtration, separation or enrichment based on electric / magnetic field; or a combination thereof and / or wherein in said step of removal of large peptides and proteins or otherbiomolecules peptides, proteins and / or other biomolecules having a molar weight larger than 20 kDa, preferably having a molar weight larger than 15 kDa, most preferably having a molar weight larger than 10 kDa are removed from the first fragment sample.

8. Method according to any of the preceding claims, wherein said perturbation is selected from the group consisting of: temperature change; pressure change; ionic strength change; pH change; metabolic stimulant change; ligand addition, including drug / small molecule addition, metabolite addition, protein addition, peptide addition, lipid addition, DNA addition, RNA addition, vitamin addition, viral entry, internalization of molecules, irradiation, disease / health state or status and genetic variations, including mutations, or a combination thereof; addition of a chaotrope; chemical modification, including post-translational modifications, in particular phosphorylations, disulfide bridge formation, ADP-ribosylation, ubiquitination, SUMOylation, acetylation, methylation, oxidation, glycosylation, or a combination thereof, phase-separation, triggers or disruptors of interactions between molecules, triggers of protein degradation,.

9. Method according to any of the preceding claims, wherein step 4. includes, before actual analysis, a proteomics workflow, in particular involving denaturation, C18 cleanup, or a combination thereof.

10. Method according to any of the preceding claims, wherein in the step 1. the proteolytic system is used at a concentration, with respect to the total biomolecular content in the sample, given as the ratio of enzyme to biomolecular content, in the range of 1 / 50- 1 / 10000, preferably in the range of 1 / 100-1 / 1000 by weight and / or wherein for quantitative determination heavy labelled fragments characteristic of being the result of the limited proteolysis of step 1., the fragmentation of step 2, or the complete fragmentation of step 3. are spiked into the first fragment sample and / or the completely fragmented fragment sample.

11. Method according to any of the preceding claims, wherein step 1. is carried out for the limited proteolysis phase over a time span of 1-60 minutes, preferably in the range of 2-30 minutes, or 2 - 10 minutes or 2 - 5 minutes, further preferably at a temperature in the range of 20-40°C.

12. Method according to any of the preceding claims, wherein in step 1. the limited proteolysis phase is terminated by quenching of protease activity, in particular byinducing conditions that denature proteins or inhibit its enzymatic activity, including an increase in temperature, transfer of the cells into media that lyse cells and denature proteins simultaneously, and / or wherein in step 1. the limited proteolysis phase is followed by a washing step, preferably using a saline solution.

13. Method according to any of the preceding claims, wherein the at least one protein is a protein based exclusively on proteinogenic amino acids, or is based on proteinogenic amino acids and carries post-translational modifications.

14. Use of a method according to any of the preceding claims for the determination, in a hypothesis-free manner, of at least one structural and / or physicochemical properties of a protein or peptide, in particular of a change of state, in particular a conformation or a change in the local or global physico-chemical properties, of said at least one protein or peptide, said at least one protein preferably having undergone a change of state, in particular conformational changes or a change in the local or global physicochemical properties, after perturbation induced in the investigated complex mixture, or of a medically relevant state, in particular conformation or local or global physico-chemical properties, of the protein, for the determination of protein-based drugs, for the influence of drugs or other ligands on proteins, or for quality control of protein-based pharmaceutical preparations.

15. Use of a method according to any of the claims 1- 13 in combination with peptide fragment enrichment techniques such as TAILS for the peptides generated by step 1.

Images