In-situ capture and identification method for protein interaction between organelle membranes in living cells

Abstract

The invention discloses a method for analyzing interaction of proteins among organelle membranes in living cells, which comprises the following steps: introducing a targeted enrichment tag on an organelle membrane, controllably activating tag reaction, realizing in-situ labeling of organelle membrane proteins, and simultaneously adopting a chemical cross-linking mode to realize in-situ labeling of organelle membrane proteins. And in-situ capture of intercellular organelle membrane protein interaction within the distance constraint range of the cross-linking agent. By integrating the two strategies, the in-situ capture and large-scale identification of the interaction of the intercellular intermembrane proteins in the living cells are realized.

Description

Technical Field

[0001] The invention relates to an in situ capture and large-scale identification method for protein interaction between organelle membranes in living cells, establishes an in situ protein interaction analysis platform between organelle membranes in living cells, and belongs to the technical field of protein interaction analysis. Background Art

[0002] The endomembrane system of eukaryotic cells compartmentalizes the intracellular space to form various organelles that are spatially independent of each other. Each organelle not only performs specific functions independently, but also exchanges substances and information between organelles through vesicle transport, membrane contact, etc. (Elie Dolgin. Nature, 2019, 567: 162-164). Protein interactions in organelle membrane-membrane-related regions are crucial to maintaining the efficient and stable operation of various physiological activities in cells. Research on the occurrence mechanism and regulatory pathways of protein interactions in organelle membrane-membrane-related regions helps to reveal the occurrence and development mechanism of diseases, and thus provide new ideas for disease diagnosis and treatment, drug development, etc. (Yiran E Li et al. Cell Death Dis., 2022, 13 (5): 504).

[0003] Classical methods for studying protein-protein interactions include yeast two-hybrid (Y2H) system, fluorescence resonance energy transfer, co-immunoprecipitation, affinity purification mass spectrometry (AP-MS) and other methods (Prabu Gnanasekaran et al. Methods Mol Biol., 2023, 2690: 81-85). Among them, the yeast two-hybrid system is suitable for preliminary screening of protein interactions, but because it detects interactions in the yeast system, the false positive rate is high. Co-immunoprecipitation is suitable for detecting interactions between protein complexes, but it has high requirements for antibodies. The AP-MS method combines mass spectrometry analysis, which is conducive to detecting low-abundance protein interactions in the complex, but it also has high requirements for antibodies.

[0004] However, none of the above methods can perform proteome-wide analysis of transient and weak interactions at the subcellular level, and cannot perform in situ capture and large-scale identification of protein interactions between organelle membranes in living cells. In recent years, proximity labeling-mass spectrometry and chemical cross-linking mass spectrometry have been established for in situ capture of transient and weak protein interactions (Wei Qin et al. Nat. Methods, 2021, 18: 133-143; Hang Gao et al. Anal. Chem., 2022, 94: 7551-7558). The enzyme-mediated proximity labeling strategy expresses ascorbate peroxidase or biotin ligase in the target area and achieves labeling by generating highly active chemical species, which can achieve in situ organelle targeted labeling. The chemical cross-linking strategy forms a covalent bond with the protein through the functional group of the small molecule cross-linker, and captures the interacting proteins within the range of the cross-linker spacer arm in situ, thereby providing information on the protein interaction interface. However, the enzyme-mediated proximity labeling strategy captures protein interactions in the area surrounding the fusion protein, and it is difficult to distinguish between direct and indirect interactions between proteins. The chemical cross-linking strategy captures protein interactions within the whole cell and lacks spatial targeting. Based on the characteristics of the above two strategies, this patent proposes a method for in situ capture and large-scale identification of protein interactions between organelle membranes in living cells using targeted enrichment labeling reaction-chemical cross-linking. Summary of the invention

[0005] In order to overcome the problems that existing protein interaction analysis methods cannot simultaneously achieve spatial targeting and protein interaction interface information analysis, the present invention combines the organelle targeting characteristics of targeted enrichment tags and the capture of protein interactions within a certain distance range by chemical cross-linkers and the analysis of interaction interface information, and provides a method for analyzing protein interactions between organelle membranes in living cells. This method has the advantages of in situ, rapid, and targeted, and can be used for the capture and large-scale identification of organelle membrane-membrane interaction proteins.

[0006] The technical solution adopted by the present invention is as follows:

[0007] (1) Cell line construction

[0008] In human 293T cell lines, ascorbate peroxidase 2 (APEX2) was fused with organelle membrane targeting peptides or organelle membrane marker proteins by constructing clones. The organelle membranes selected for fusion included the endoplasmic reticulum membrane (ERM), outer mitochondrial membrane (OMM) and plasma membrane (PM).

[0009] (2) Organelle-targeted in situ labeling and chemical cross-linking

[0010] In the cell line fused with APEX2 tag, biotin-phenol is used as substrate and hydrogen peroxide solution is used as activator to controllably activate the biotinylation reaction of proteins around the organelle membrane of the APEX2 tag fusion region. Then, chemical crosslinkers are added to perform whole-cell in situ crosslinking reaction to covalently link the interacting proteins in the targeted labeled region within the distance constraint of the crosslinker.

[0011] (3) Cell lysis and protein extraction:

[0012] Add 0.1-10 ml of protein extraction solution (one or more of SDS, Triton X-100, RapiGest SF or NP-40, with a mass volume concentration of 0.1%-20% based on the ratio of the mass of the surfactant (in g) to the volume of the buffer solution (in mL)), and perform ultrasound-assisted cell lysis. The ultrasound conditions are 40-120 W, 1-10 min, and 0-4°C. Use one or more of the protein precipitation reagents methanol, chloroform, acetone, etc., with a volume of 4-8 times the protein extraction solution, to precipitate the protein at -20°C for 6-24 hours, and remove excess small molecule reagents.

[0013] (4) Enrichment reaction of labeled proteins

[0014] Add agarose gel ball enrichment material with streptavidin as the monomer to the product of step (3) to enrich the biotinylated protein and obtain the cross-linked protein in the target labeling area.

[0015] (5) Proteomics sample preparation and large-scale identification

[0016] Combined with the proteomics sample preparation method, the protein enriched in step (4) is denatured, reduced, alkylated and enzymatically hydrolyzed to obtain cross-linked peptides for LC-MS / MS analysis.

[0017] (6) Cross-linking data analysis

[0018] The mass spectrometry data were retrieved using Plink 2.0.

[0019] This method was used to analyze cell lines targeting two organelle membranes. The cross-linked protein pairs identified in both organelle membranes were the interacting proteins in the region between the two organelle membranes. By analyzing cell lines targeting multiple organelle membranes, such as the mitochondrial outer membrane, endoplasmic reticulum membrane, plasma membrane, and nuclear membrane, an in situ protein interaction analysis platform between organelle membranes in living cells was established, achieving in situ capture and large-scale identification of protein interactions between organelle membranes in living cells.

[0020] The present invention has the following advantages:

[0021] 1. The method integrates two strategies: organelle-targeted in situ labeling and chemical cross-linking, to achieve in situ capture of protein-protein interactions between organelle membranes in living cells.

[0022] 2. The method introduces targeted enrichment tags on the organelle membrane for in situ labeling of organelle membrane proteins, and uses chemical cross-linking to capture protein interactions within the distance constraint of the cross-linker in situ, which can achieve accurate capture of protein interactions between organelle membranes.

[0023] 3. Combining proteomics and mass spectrometry analysis methods can achieve large-scale identification of protein interactions between organelle membranes. DETAILED DESCRIPTION

[0024] This embodiment is implemented on the premise of the technical solution of the present invention, and provides a detailed implementation method and a specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0025] Example 1

[0026] The technical solution adopted by the present invention is as follows:

[0027] (1) Organelle-targeted in situ labeling

[0028] HEK293T cell lines expressing ascorbate peroxidase on the outer membrane of mitochondria (APXE2-OMM) and HEK293T cell lines expressing ascorbate peroxidase on the endoplasmic reticulum membrane (APXE2-ERM) were selected and cultured in a 10 cm inner diameter culture dish to a density of 80%-90% (adherent cells occupy approximately 80%-90% of the culture dish area) (Ying Zhou et al. Angew. Chem., Int. Ed. Engl., 2019, 58, 11763-11767; Yuxin An et al. al.Anal.Chem., 2022, 94, 12051-12059); Prepare a 0.5M biotin phenol (BP) stock solution (Sigma-Aldrich, SML2135) with DMSO, dilute the BP stock solution to 0.5mM with DMEM culture medium preheated at 37°C, take 4.5mL and add it to the cells in a culture dish with an inner diameter of 10cm, incubate at 37°C and 5% CO2 in air by volume for 30min.

[0029] Add 100 mM hydrogen peroxide solution at a volume ratio of 100:1 to the cell culture medium, mix well, and let stand for 1 min; immediately discard the supernatant, add 8 mL of quenching solution (buffered saline solution containing 10 mM sodium ascorbate, 5 mM Trolox, and 10 mM sodium azide, all purchased from Sigma-Aldrich), let stand for 2 min, discard the quenching solution, and add 5 mL of 10 mM phosphate buffered saline (pH = 7.4) to wash once.

[0030] (2) Chemical cross-linking reaction:

[0031] Prepare a 0.5M cross-linker DSDHD stock solution with DMSO (Yuxin An et al. Anal. Chem., 2022, 94, 12051-12059), dilute the stock solution with phosphate buffer to a final concentration of 5mM, take 4.2mL and add it to the cells, incubate at room temperature for 10min, and discard the solution. Add 1mL of phosphate buffer, scrape the cells from the dish with a cell scraper, and add ammonium bicarbonate solution with a final concentration of 50mM (the stock solution concentration is 1mol / L) to terminate the reaction for 10min.

[0032] (3) Cell lysis and protein extraction:

[0033] Add 1 mL of 4% sodium dodecyl sulfate (SDS)-phosphate buffer solution (phosphate buffer solution concentration is 10 mM, pH = 7.4), and perform cell lysis with ultrasound on ice. The ultrasound conditions are 80W, 5s on / off, 5min. Use 8 mL of pre-cooled acetone at -20°C for protein precipitation for 12h, remove excess small molecule reagents, then centrifuge to remove acetone, and evaporate to obtain protein precipitation.

[0034] (4) Enrichment of target protein:

[0035] Use a 1% mass concentration SDS solution to redissolve the protein precipitate in step (4). After complete dissolution, add phosphate buffer to make the final mass concentration of SDS 0.3%. Add 250uL of agarose gel beads enrichment material with streptavidin (Thermo Fisher Scientific, 20353), incubate at room temperature for 3h, and enrich the target protein. After enrichment, use a 1% mass concentration SDS solution, 8M urea solution, and 2M sodium chloride solution to wash the enrichment material twice each.

[0036] (5) Proteomics sample preparation and mass spectrometry identification

[0037] Add 6M urea-ammonium bicarbonate solution (ammonium bicarbonate solution concentration is 50mM) to the enriched material obtained in step (4), add dithiothreitol (DTT) to a final concentration of 10mM, react at room temperature for 1h to denature and reduce the protein; then add iodoacetic acid (IAA) to a final concentration of 20mM, react at room temperature in the dark for 30min to alkylate the protein; after the reaction, wash the enriched material 3 times with 50mM ammonium bicarbonate, then add trypsin solution to the sample at a mass ratio of protein:trypsin = 50:1, and react at 37°C for 16h.

[0038] The supernatant was collected by centrifugation, and then a 0.1% FA-water solution was added to wash the enriched material. After the supernatant was collected by centrifugation, the two supernatant solutions were combined to obtain the target peptide solution; the peptide solution was desalted and high pH reverse phase fractionated, and each fraction was lyophilized and re-dissolved for mass spectrometry analysis. Desalting and high pH reverse phase fractionation were performed using a Waters UPLC system equipped with a WatersXSelect Peptide CSH C18 column (specification 2.1mm×150mm, filler particle size 3.5μm). The peptide sample was dissolved in 5mM ammonium bicarbonate solution (pH=10.0) and then loaded onto the chromatographic column for desalting and high pH reverse phase fractionation. The mobile phase A phase composition was 5mM ammonium bicarbonate solution (pH=10.0), and the phase B composition was 90% acetonitrile-10% 5mM ammonium bicarbonate solution (pH=10.0) by volume. Desalting was performed using a 2% B (volume ratio) phase and run at a flow rate of 0.2mL / min for 15min. Afterwards, high pH reverse phase fractionation was performed, and the composition of phase B (volume ratio, the same below) was gradually eluted from low to high to elute the peptides: 0-10min, 2% phase B; 10-11.5min increased to 10% phase B, 11.5-33.5min increased to 36% phase B, 33.5-38.5min increased to 65% phase B, 38.5-40min increased to 90% phase B, 40-48min maintained at 90% phase B, 48min decreased to 2% phase B until 50min ended. 10-40min was the effective gradient time period, and one fraction was collected every 1min, a total of 30 fractions were collected, and they were merged into 10 fractions (fractions of 10-11min, 20-21min and 30-31min were merged, and fractions of 11-12min, 21-22min and 31-32min were merged, and the rest were deduced by analogy).

[0039] (6) Cross-linking data retrieval and analysis

[0040] The generated data files were searched using pLink 2.0, and the Human reviewed FASTA database (2022-11-11) in the Uniprot website was used for retrieval. Among them, 1074 cross-linked peptides identified in the APEX2-OMM and APEX-ERM datasets corresponded to 437 pairs of protein interactions of 513 proteins, corresponding to 869 protein interaction sites. The 513 proteins identified included proteins such as UBB and PHB2 that have mitochondrial localization in the Uniprot GO annotation, and proteins such as ERLIN1 and ERLIN2 that have endoplasmic reticulum localization in the Uniprot GO annotation.

[0041] Example 2

[0042] The steps (process and conditions) are the same as in Example 1, except that: HEK293T cell lines expressing ascorbate peroxidase (APXE2-OMM) on the outer membrane of mitochondria and HEK293T cell lines expressing ascorbate peroxidase (APXE2-PM) on the cell membrane are selected for treatment (Ying Zhou et al. Angew. Chem., Int. Ed. Engl., 2019, 58, 11763-11767; Yuxin An et al. Anal. Chem., 2022, 94, 12051-12059).

[0043] pLink 2.0 was used to search the generated data files. Among them, 757 cross-linked peptides identified in the APEX2-OMM and APEX-PM datasets corresponded to 229 pairs of protein interactions of 304 proteins, corresponding to 578 protein interaction sites. The 304 proteins identified included proteins with mitochondrial localization in Uniprot GO annotations such as UBB and PHB2, and proteins with plasma membrane localization in Uniprot GO annotations such as PHB1.

[0044] Example 3

[0045] The steps (process and conditions) are the same as Example 1, except that: a human 293T cell line expressing ascorbate peroxidase (APXE2-ERM) on the endoplasmic reticulum membrane and a human 293T cell line expressing ascorbate peroxidase (APXE2-PM) on the cell membrane are selected for treatment (Ying Zhou et al. Angew. Chem., Int. Ed. Engl., 2019, 58, 11763-11767; Yuxin An et al. Anal. Chem., 2022, 94, 12051-12059).

[0046] pLink 2.0 was used to search the generated data files. Among them, 728 cross-linked peptides identified in the APEX2-OMM and APEX-PM datasets corresponded to 229 pairs of protein interactions of 294 proteins, corresponding to 650 protein interaction sites. The 294 proteins identified included LRC59, HSPA1A and other proteins with endoplasmic reticulum localization in Uniprot GO annotation, and XPC and other proteins with plasma membrane localization in Uniprot GO annotation.

[0047] Example 4

[0048] The steps are the same as (process and conditions) as in Example 1, except that: human 293T cell lines expressing ascorbate peroxidase (APXE2-ERM) on the endoplasmic reticulum membrane and human 293T cell lines expressing ascorbate peroxidase (APXE2-NM) on the nuclear membrane are selected for treatment (Ying Zhou et al. Angew. Chem., Int. Ed. Engl., 2019, 58, 11763-11767; Yuxin An et al. Anal. Chem., 2022, 94, 12051-12059).

[0049] pLink 2.0 was used to search the generated data files. Among them, 714 cross-linked peptides identified in the APEX2-ERM and APEX-PM datasets corresponded to 212 pairs of protein interactions of 286 proteins, corresponding to 623 protein interaction sites. The 286 proteins identified included proteins such as CTTN that were localized in the endoplasmic reticulum in the Uniprot GO annotation, and proteins such as PCBP3 and CUX1 that were localized in the nucleus in the Uniprot GO annotation.

[0050] Example 5

[0051] The steps are the same as (process and conditions) as in Example 1, except that: human 293T cell lines expressing ascorbate peroxidase (APXE2-OMM) on the outer membrane of mitochondria and human 293T cell lines expressing ascorbate peroxidase (APXE2-NM) on the nuclear membrane are selected for treatment (Ying Zhou et al. Angew. Chem., Int. Ed. Engl., 2019, 58, 11763-11767; Yuxin An et al. Anal. Chem., 2022, 94, 12051-12059).

[0052] pLink 2.0 was used to search the generated data files. Among them, 698 cross-linked peptides identified in the APEX2-OMM and APEX-NM datasets corresponded to 194 pairs of protein interactions of 265 proteins, corresponding to 607 protein interaction sites. The 265 proteins identified included proteins with mitochondrial localization in Uniprot GO annotations such as PPP2CA, and proteins with nuclear localization in Uniprot GO annotations such as HNRPC and HNRC1.

[0053] Example 6

[0054] The steps (process and conditions) are the same as those in Example 1, except that: human 293T cell lines expressing ascorbate peroxidase (APXE2-OMM) on the outer membrane of mitochondria, ascorbate peroxidase (APXE2-ERM) on the endoplasmic reticulum membrane, and human 293T cell lines expressing ascorbate peroxidase (APXE2-PM) on the cell membrane are selected for treatment (Ying Zhou et al. Angew. Chem., Int. Ed. Engl., 2019, 58, 11763-11767; Yuxin An et al. Anal. Chem., 2022, 94, 12051-12059).

[0055] The generated data files were searched using pLink 2.0. Among them, 546 cross-linked peptides identified in the APEX2-OMM, APEX-ERM, and APEX-PM datasets corresponded to 122 pairs of protein interactions of 182 proteins, corresponding to 596 protein interaction sites. The 182 proteins identified included proteins such as HSP90B that have mitochondrial localization in the Uniprot GO annotation, and proteins such as HSPA1A that have nuclear localization in the Uniprot GO annotation.

Claims

1. A method for in situ capture and identification of protein interactions between organelle membranes in living cells, characterized by: Targeted enrichment tags are introduced on the organelle membrane, and targeted in situ labeling of proteins between organelle membranes is achieved through controllable activation of the targeted tags. At the same time, a chemical cross-linking strategy is adopted to covalently link interacting proteins within the range of the cross-linker spacer arm length to capture protein interactions in situ, and finally the protein cross-linking information in the targeted area is obtained by enriching the targeted tags.

2. The identification method according to claim 1, characterized in that: The method integrates the organelle targeted labeling strategy and the chemical cross-linking strategy, and uses a chemical cross-linking agent to covalently link the interacting proteins in the targeted region while targeting in situ labeling of proteins in the inter-membrane region of the organelle; This method is used to analyze the protein interactions in the membrane regions of more than two different organelles. The cell lines that target more than two different organelle membranes (each cell line carries an organelle membrane targeting enrichment tag) are respectively targeted with the same targeting enrichment tag for in situ cell targeting labeling, and chemically cross-linked with the same cross-linking agent. The tagged proteins are then enriched and proteomic analysis is performed. The cross-linked protein pairs identified by the cell lines that target two or more different organelle membranes are the interacting proteins in the corresponding organelle membrane regions. By analyzing and verifying these interacting proteins, the in situ capture and large-scale identification of protein interactions between organelle membranes in living cells are finally achieved.

3. The identification method according to claim 1, characterized in that: In the cell line, the targeted enrichment tag is fused with the organelle targeting peptide or organelle marker protein by constructing a clone. The corresponding substrate is selected according to the tag type to activate the targeted tag reaction. The cell line is one or more of human embryonic kidney cells (HEK293T), human cervical cancer cells (Hela), and human neuroblastoma cells (SH-SY5Y); The organelle targeting peptide or organelle marker protein is one or more of the endoplasmic reticulum membrane marker protein Calnexin, the mitochondrial outer membrane marker protein TOMM20, the nuclear localization signal sequence (NLS), and the cell membrane localization signal sequence CAAX.

4. The identification method according to claim 1 or 2, characterized in that: Targeted enrichment tag types include: one or more of the Escherichia coli biotin ligase BirA mutant (BioID, TurboID), ascorbate peroxidase (APEX), and mini-singlet oxygen producer (miniSOG) proteins; The substrate includes one or more of biotin-phenol and biotin, and the activator includes one or more of hydrogen peroxide, rapamycin, and light activation; The light excitation wavelength is the wavelength required by the photoreactive group, such as 254 nm or 365 nm; the light treatment time is 10 s-10 min.

5. The identification method according to claim 1, characterized in that: While targeting the labeling of proteins in the inter-membrane region of organelles, a chemical cross-linking strategy was used to covalently link interacting proteins within the distance constraints of the cross-linker to capture protein interactions in situ; The chemical crosslinking agent is an organic substance with reactive groups at both ends of a C3-C20 alkyl chain, which has good membrane permeability and can achieve a high coverage of live cell crosslinking reaction; the crosslinking agent arm length range is The types of groups include: N-hydroxysuccinimide (NHS) groups and imidoesters that react with primary amines (-NH2); maleimide groups and pyridyldithiol groups that react with sulfhydryl groups (-SH); carbodiimide and acetyl-N-hydroxysuccinimide that react with carboxyl groups (-COOH); The specific cross-linking agent is one or more of bissuccinimidyl sulfoxide (DSSO, Thermo Fisher Scientific, A33545) and 6,7-dihydroxydodecanedioate DSDHD (Yuxin An et al. Anal. Chem., 2022, 94, 12051-12059).

6. The method according to claim 1, characterized in that: The enrichment methods of the targeted tags include one or more of affinity capture, immunoprecipitation, tag affinity enrichment, etc.; The enrichment material is a material with an enrichment group on the matrix, and the enrichment group is one or more of avidin, streptavidin, etc. that can bind with high specificity to the targeted enrichment tag; The matrix of the enrichment material is one or more organic / inorganic materials such as agarose gel balls, magnetic balls, PEG polymer balls, etc.

7. The identification method according to claim 1, characterized in that: The organelle membrane types include one or more of the endoplasmic reticulum membrane (ERM), the outer mitochondrial membrane (OMM), the nuclear membrane (NM), the plasma membrane (PM), etc.; Two of the cell lines with targeted organelle membranes were analyzed, and the cross-linked protein pairs identified in both organelle membranes were the interacting proteins in the region between the two organelle membranes. By analyzing different organelle membranes, the in situ capture and large-scale identification of protein interactions between organelle membranes in living cells were achieved.

8. The identification method according to claim 1, characterized in that: Targeted enrichment tags are introduced on the organelle membrane, and targeted in situ labeling of proteins between organelle membranes is achieved through controllable activation of targeted tags. At the same time, a chemical cross-linking strategy is adopted to covalently link interacting proteins within the distance constraint of the cross-linker to capture protein interactions in situ. The labeled proteins are enriched to obtain cross-linked proteins in the targeted labeling area, and the proteins are processed in combination with proteomics sample preparation technology. They are identified and cross-linked data are analyzed through high performance liquid chromatography-mass spectrometry technology, and an in situ protein interaction analysis platform between organelle membranes in living cells is established to generate a protein dataset for organelle membrane interactions in living cells.

9. The identification method according to any one of claims 1 to 8, characterized in that: This method was used to analyze different organelle membranes, such as the mitochondrial outer membrane and the endoplasmic reticulum membrane. The cross-linked proteins identified in both datasets were the interacting proteins in the region between the mitochondrial outer membrane and the endoplasmic reticulum membrane. The region between the mitochondrial outer membrane and the endoplasmic reticulum membrane, namely the mitochondria-associated endoplasmic reticulum membrane region (MAM), is involved in the bidirectional regulation of lipid and Ca 2+ Important physiological processes such as homeostasis, mitochondrial dynamics, autophagy and apoptosis; By using this method to capture and identify the interacting proteins in this region on a large scale in situ, it is expected that the establishment of a dataset of interacting proteins in the MAM region can be achieved, thereby providing data support for the study of various physiological and pathological processes in this region.