key: cord-1056550-deql0cgj
authors: V’kovski, Philip; Steiner, Silvio; Thiel, Volker
title: Proximity Labeling for the Identification of Coronavirus–Host Protein Interactions
date: 2020-05-11
journal: Coronaviruses
DOI: 10.1007/978-1-0716-0900-2_14
sha: 0341aef4c5f4ac60f2a63af2b8d205b438f3951a
doc_id: 1056550
cord_uid: deql0cgj

Biotin-based proximity labeling circumvents major pitfalls of classical biochemical approaches to identify protein–protein interactions. It consists of enzyme-catalyzed biotin tags ubiquitously apposed on proteins located in close proximity of the labeling enzyme, followed by affinity purification and identification of biotinylated proteins by mass spectrometry. Here we outline the methods by which the molecular microenvironment of the coronavirus replicase/transcriptase complex (RTC), i.e., proteins located within a close perimeter of the RTC, can be determined by different proximity labeling approaches using BirA(R118G) (BioID), TurboID, and APEX2. These factors represent a molecular signature of coronavirus RTCs and likely contribute to the viral life cycle, thereby constituting attractive targets for the development of antiviral intervention strategies.

Coronaviruses cause a wide range of diseases in animals and humans. In recent years, particular attention has been drawn by severe acute respiratory syndrome coronavirus (SARS-CoV), Mid-dleEast respiratory syndrome coronavirus (MERS-CoV) and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which can cause severe and lethal respiratory disease in humans [1] [2] [3] [4] [5] . Moreover, in the veterinary field, coronaviruses such as, feline infectious peritonitis virus, porcine epidemic diarrhea virus, or the avian infectious bronchitis virus, severely affect companion animals and livestock [6] [7] [8] .

Viral pathogenicity is determined by the virus tissue and cell tropism and the host (inflammatory) responses to the infection, as well as viral countermeasures evading the host immune defense system.

On the cellular level, the expression of up to 15-16 non-structural proteins carrying various enzymatic functions result in the establishment of viral replication and transcription complexes (RTC) and the generation of membranous structures that host the RTC in the infected host cell cytosol [9, 10] . This stage is considered a major determinant of pathogenicity and infection outcome. Indeed, host proteins may be recruited to the RTC to promote viral replication [11, 12] or may display antiviral functions to restrict virus replication [13] . Knowledge on the composition of viral and host proteins located at the RTC is therefore of crucial importance to understand critical virus-host interactions taking place at the site of viral RNA synthesis.

Proximity labeling approaches have been implemented in an increasing number of investigations during the past years [14] . The strong enthusiasm of the scientific community is best exemplified by the constant improvement and adaptation of proximity labeling enzymes to a wide range of applications [15] .

The hallmark of enzyme-catalyzed proximity labeling is the promiscuous and covalent biotin labeling of proteins located within a close perimeter (few nanometers). As such, the labeling is not dependent on the protein-protein interaction affinity. The high affinity between biotinylated residues and streptavidin, however, allows stringent and efficient affinity purification using streptavidin-coated beads. Eventually, affinity-purified proteins are sensitively identified by mass spectrometry. Additionally, proximity labeling is also suited to capture transient interactions, as the labeling time can vary from 1 min to several hours depending on the enzyme [15] [16] [17] [18] . Of note, the labeling by most enzymes is not toxic and can be performed in live cells.

In this context, BirA R118G (BioID) was the first proximity labeling enzyme successfully employed in different assays [16] . Bir-A R118G is a promiscuous E. coli biotin ligase that uses free biotin in labeling reactions that require several hours of biotin incubation to obtain sufficient amounts of biotinylated proteins for further analysis. During viral infections, this feature can be advantageous when recording protein-protein interactions occurring at any stage during the entire course of infection in an unbiased screening approach [19] .

The slow labeling kinetics of BioID were first improved by the development of BioID2 [18] . Recently, Branon et al. used directed evolution of BioID to engineer TurboID and miniTurbo, which contained few key amino acid substitutions conferring similar labeling capacities within 10 min instead of the 15-18 h required by BioID [15] . This significant improvement allows retaining the intrinsic advantages of proximity labeling enzymes while narrowing the labeling window. When applied to virus infections, TurboID enables to distinguish protein-protein interactions relevant at defined stages of infection.

Another popular proximity labeling enzyme is the soy beanderived ascorbate peroxidase APEX2 [17, 20] . APEX2 catalyzes biotin-phenol substrates into biotin-phenoxyl radicals that react and tag neighboring proteins. This reaction is triggered by the addition of hydrogen peroxide to cells for 1 min. APEX2 can thereby be used to provide a "snapshot" of factors surrounding the protein of interest to which it is fused.

Alternatively, APEX2 catalyzes the polymerization of 3,3-diaminobenzidine (DAB) resulting in the deposition of insoluble DAB polymers at the site of production [21] . Given that DAB precipitates generate high contrast upon heavy-metal staining, this feature allows detecting, in a specific manner, the localization of an APEX2 fusion protein by electron microscopy.

Lastly, APEX2 has been demonstrated to label closely associated RNAs [22] . Upon purification of biotinylated RNAs, sequencing reveals the RNA population associated with a particular protein complex.

Here, we describe the use of BioID, TurboID, and APEX2 in the context of viral infections (Fig. 1 ). More specifically, by incorporating these proximity labeling enzymes into the RTC of a prototype coronavirus, these strategies enable the identification of critical host factors comprised within the coronavirus RTC microenvironment. The proximity labeling procedures described here are largely adapted from detailed protocols available for BioID [23] , Fig. 1 Overview of proximity labeling assays using MHV-BirA R1118G -nsp2, MHV-TurboID-nsp2, and MHV-A-PEX2-nsp2

TurboID [15] , and APEX2 [17, 21] . Further procedures such as mass spectrometry of affinity purified proteins, electron microscopy, or data analysis are not covered in this chapter. 13. L929 cells. [26] . Coding sequences for BirA R118G , TurboID and APEX2 are derived from Addgene cat. no. 74223, 116904, and 72480, respectively, and cloned into a pGPT-1 vector as described previously [19] . Recombinant MHV viruses are generated using an established vaccinia virus-based reverse genetic system [27, 28] . N-terminally tagged proximity labeling enzymes are inserted in the MHV genome within ORF1a. Note the preserved polyprotein cleavage sites ensuring the release of the nsp2 fusion protein (Fig. 2, black arrows) . Additionally, a flexible (SGG) 3 linker is placed between proximity labeling enzymes and nsp2 to provide structural flexibility and avoid potential steric hindrance of the fusion protein when embedded in the MHV RTC (see Note 2). ORF1a nsp1, nsp2, and nsp3 are represented by gray boxes; BirA R118G (BioID), TurboID, and APEX2 by a dark gray box; and molecular V5 and myc tag by a black box. The amino acid and nucleotide sequence at the junction between nsp1 and the tagged proximity labeling enzyme are highlighted. The junction between the proximity labeling enzyme and nsp2, in which a flexible linker is incorporated is highlighted as well. Black arrows indicate proteolytic cleavage sites 5. Establish that a functional fusion protein is expressed at the desired subcellular location by performing immunofluorescence microscopy and colocalization analyses and by detecting biotinylated proteins by both immunofluorescence microscopy and western blot analysis (see Subheadings 3.6 and 3.7). 5. Proceed by fixating the cell for immunofluorescence microscopy analysis or prepare lysates for western blot analysis as described in Subheadings 3.6 and 3.7. 6. Proceed by fixating the cell for immunofluorescence microscopy analysis or prepare lysates for western blot analysis as described in Subheadings 3.6 and 3.7. 14. For display purposes, adjust brightness and contrast to appropriate controls during post-acquisition processing of immunofluorescence microscopy images.

Analysis of Biotinylated Proteins (Analytical Scale)

1. Perform infections and labeling procedures for analytical western blots in six-well cell culture plates as indicated in Subheadings 3.2, 3.3, or 3.4.

Prepare lysates using M-PER mammalian protein extraction reagent, RIPA, or other lysis buffers (see Subheadings 3.9-3.11) containing protease inhibitors. Typically, 100-200 μl/ well lysis buffer is used, and cells are scraped off using an inverted pipette tip.

3. Collect lysate in 1.5 ml tube, centrifuge at~14,000 Â g for 5-10 min to pellet the cell debris and transfer supernatant to new tube. Add Laemmli buffer and boil for 5-10 min.

4. To assess biotinylated proteins in affinity-purified fractions, see Subheadings 3.9-3.11.

5. Separate proteins on a 10% (w/v) SDS-polyacrylamide gel.

6. Electroblot onto a nitrocellulose membrane.

7. Incubate nitrocellulose membrane in protein-free blocking buffer for 60 min at room temperature (see Note 7).

8. Incubate nitrocellulose membrane with horseradish peroxidase (HRP)-conjugated streptavidin diluted in protein-free blocking buffer. Incubate at 4 C overnight.

9. Wash the nitrocellulose membrane three times with PBS supplemented with 0.5% (v/v) Tween 20 and once with PBS.

10. Visualize biotinylated proteins using an enhanced chemiluminescence (ECL) HRP substrate and a chemiluminescence CCD detector system. 

5. Keep the cells on ice for the rest of the procedure.

6. Pool the lysates of the four flasks into one 50 ml conical tube.

7. Add 400 μl 20% Triton X-100 to each tube.

Sonifier 250 (at 30% constant, 30% power). Put cells on ice during the sonication rounds. 11. Transfer lysates to 4 Â 2 ml tubes and centrifuge 10 min at 18,000 Â g, 4 C.

12. Wash streptavidin-coated magnetic beads (800 μl beads per condition) with 1 ml lysis buffer diluted 1:1 with 50 mM Tris-HCl pH 7.4. Place the cells on a magnetic tube holder and remove the wash solution.

13. Repeat wash with 1 ml of lysis buffer diluted 1:1 with 50 mM Tris-HCl pH 7.4, and distribute the solution to four 2 ml tubes. Remove wash solution using a magnetic tube holder.

14. Take an aliquot of 100 μl from the lysate (total 8 ml) before incubation with the magnetic beads as control and add Laemmli buffer. Boil 10 min at 98 C and store at À20 C for further western blot analysis.

15. Distribute the remaining lysate to the four 2 ml tubes containing the washed streptavidin-coated beads and incubate on a rotator at 15 rotations per minute at 4 C overnight.

16. Place the samples on a magnetic tube holder and collect the "flowthrough" (non-bound protein lysate). Combine the beads into one tube.

17. Take an aliquot of 100 μl from the flowthrough and add Laemmli buffer. Boil 10 min at 98 C and store at À20 C for further western blot analysis.

18. Wash the beads as described below. After each washing step, place the tube on the magnetic tube holder and remove the washing solution.

19. Wash beads twice with 1.5 ml 2% (w/v) SDS by incubating beads for 5-8 min on a rotator.

20. Wash once with 1.5 ml BioID washing buffer 2 by incubating beads for 5-8 min on a rotator.

Wash once with 1.5 ml BioID washing buffer 3 by incubating beads for 5-8 min on a rotator. 22 . Wash once with 1 ml 50 mM Tris-HCl pH 7.4.

Laemmli SDS-sample buffer supplemented with 0.5 mM biotin and heating at 95 C for 10 min in a heating block while shaking at 700 rpm.

24. Keep 10% of eluate for western blot analysis.

25. Refer to the mass spectrometry facility's guidelines for sample submission. Samples are typically separated 1 cm into an SDS-polyacrylamide gel, stained with Coomassie, and extracted from the gel for mass spectrometry analysis (see Note 13). 7. Lyse the cells in 1.5 ml RIPA buffer containing protease inhibitors. After addition of RIPA buffer, incubate for 1-2 min before using a 1 ml pipette to detach the cells from the surface of the dish.

8. Transfer the lysate to 2 ml tubes and spin at 12,000 Â g, 4 C.

9. Using a magnetic tube holder, wash streptavidin-coated magnetic beads twice with 1 ml RIPA buffer. Use 350 μl beads per condition (see Note 15).

10. Take a 100 μl aliquot of the lysate before incubation with the magnetic beads as control and add Laemmli buffer. Boil 10 min at 98 C and store at À20 C for further western blot analysis.

11. Add the lysate to the streptavidin-coated beads and incubate on a rotator at 4 C overnight.

12. Place the samples on a magnetic tube holder and collect the "flowthrough" (non-bound protein lysate).

13. Take an aliquot of 100 μl from the flowthrough and add Laemmli buffer. Boil 10 min at 98 C and store at À20 C for further western blot analysis.

14. Wash the beads as described below. After each washing step, place the tube on the magnetic tube holder and remove the washing solution.

15. Wash the beads twice with 1 ml RIPA buffer for 2-5 min.

16. Wash the beads with 1 ml 1 M KCl for 2-5 min.

17. Wash the beads with 1 ml 0.1 M Na 2 CO 3 . Homogenize the beads with a 1 ml pipette and continue immediately with next washing step.

18. Wash the beads with 1 ml 2 M urea, 10 mM Tris-HCl (pH 8.0). Homogenize the beads with a 1 ml pipette and continue immediately with next washing step.

19. Wash the beads with 1 ml 10 mM Tris-HCl (pH 8.0).

Laemmli SDS-sample buffer supplemented with 0.5 mM biotin and heating at 95 C for 10 min in a heating block while shaking at 700 rpm.

21. Place the beads on the magnetic tube holder and collect eluates in fresh 1.5 ml tubes.

22. Refer to the mass spectrometry facility's guidelines for sample submission. Samples are typically separated 1 cm into an SDSpolyacrylamide gel, stained with Coomassie and extracted from the gel for mass spectrometry analysis (see Note 13). 13. Transfer samples into a 2 ml tube and centrifuge at 16,000 Â g at 4 C for 10 min.

14. Transfer the supernatant into a fresh 2 ml tube. This is considered the Lysate (see Note 19) .

15. Prepare 100 μl streptavidin-coated magnetic beads per sample in a 2 ml tube. 16 . Add 1 ml of APEX lysis buffer to the beads and place the tube on a rotator at 15 rotations per minute for 8 min.

17. Place the tube on a magnetic tube holder for 1 min and subsequently remove all liquid from the tube without touching the beads on the wall.

18. Repeat the washing step again for a total of two washes with APEX lysis buffer. 19 . Take an aliquot of 100 μl from the lysate and add Laemmli buffer. Boil 10 min at 98 C and store at À20 C for further western blot analysis. 20 . Add the remaining lysate to the beads and incubate the samples on a rotator at 15 rotations per minute at 4 C overnight.

21. Place the samples on a magnetic tube holder and collect the "flowthrough" (non-bound protein lysate). 22 . Take an aliquot of 100 μl from the flowthrough and add Laemmli buffer. Boil 10 min at 98 C and store at À20 C for further western blot analysis.

23. Subject the beads to a series of washing steps, described below. Each step consists of 1 ml of the respective washing buffer followed by 8 min on the rotator and 1 min on magnetic tube holder. Always remove the wash buffers before the next washing step. 3. One unique band, which is migrating at the expected molecular weight, is expected. 5. Be careful not to detach too many cells during the washing. Remove the remaining liquid from the washes.

6. Alternatively, coverslips can be left in 24-well plate. In this case, dilute antibodies in a sufficient volume to cover coverslips.

7. Milk is to be avoided as it contains biotin, which competitively inhibits biotinylated protein recognition by HRP-coupled streptavidin [29] .

8. Increase MOI if necessary in order to obtain 70-90% of infected cells.

9. Prepare 10 ml using a 10Â DAB stock. Prepare fresh before use and discard unused buffer. 10 . Add H 2 O 2 to the buffer as a last step and just before use. Also see Note 1.

11. Do not wash the cells after infection. This will allow the cells to be infected in a non-synchronized fashion and enable to record cells during different stages of the viral life cycle at the moment of lysis. Also see Note 4.

12. Repeated washing is important to completely remove free biotin, which might interfere with binding of proteins to streptavidin-coated beads.

13. Alternatively, proteins are not eluted from the beads and are submitted to on-bead tryptic digestion for mass spectrometry analysis.

14. If precise synchronization of the virus infection is desired, infect the cells at 4 C and subsequently incubate them 60-90 min at 4 C. This allows the virus particles to attach to the receptor but prevents entry. Afterwards place the cells in the incubator at 37 C. Add 25 mM HEPES to the infection medium to buffer the cells.

15. The use of streptavidin-coated magnetic beads from Pierce are recommended when using the washing conditions described in Subheading 3.10. The use of MyOne Streptavidin C1 beads has proven to result in clumping and strong adherence to the walls of tubes, thus rendering washing and elution steps difficult. Lysis and affinity purification of biotinylated proteins in MHV-TurboID-nsp2-infected cells can alternatively be performed using the conditions described in Subheading 3.9. Nevertheless, most recent publications favor the use of buffers described in Subheading 3.10.

16. Prepare one extra 10 cm dish, which can be used to count the cells and calculate the MOI for the others.

17. Be careful not to detach too many cells during the washing. Remove the remaining liquid from the washes.

18. If you need to process multiple samples, store the falcon tubes containing the lysates on ice in the meantime. 19 . The lysate can also be frozen at À80 C until further processing.

20. Eluates can be immediately processed or frozen at À80 C for storage.

Seed 2 Â 10 6 L929 cells in a 10 cm 2 dish. Cells are cultured in MEM+/+. Incubate at 37 C, 5% CO 2 for 16-18 h. 2. Infect the cells with MHV-APEX2-nsp2 or MHV-WT at MOI ¼ 4 (see Note 16)

After 1 h remove the inoculum and wash the cells three times with PBS. Add fresh MEM+/+ (see Note 12)

Thirty minutes before the desired time point, exchange the medium with 10 ml MEM+/+ supplemented with 500 μM biotin-phenol (BP)

At the desired time-point, add 100 μl H 2 O 2 solution (100Â) directly to the dish containing the 10 ml MEM+/+ supplemented with BP. Shake gently to mix

Incubate for exactly 1 min at RT

Quickly aspirate the labeling solution and wash the cells immediately three times with 10 ml of Quencher solution (see Note

Isolation of a novel coronavirus from a man with pneumonia in Saudi Arabia

Identification of a novel coronavirus in patients with severe acute respiratory syndrome

SARS and MERS: recent insights into emerging coronaviruses

A decade after SARS: strategies for controlling emerging coronaviruses

An update on feline infectious peritonitis: virology and immunopathogenesis

Porcine epidemic diarrhea virus: an emerging and re-emerging epizootic swine virus

Coronavirus avian infectious bronchitis virus

SARS-coronavirus replication is supported by a reticulovesicular network of modified endoplasmic reticulum

Infectious bronchitis virus generates spherules from zippered endoplasmic reticulum membranes

Genome-wide screen reveals valosincontaining protein requirement for coronavirus exit from endosomes

A kinome-wide small interfering RNA screen identifies proviral and antiviral host factors in severe acute respiratory syndrome coronavirus replication, including double-stranded RNA-activated protein kinase and early secretory pathway proteins

Human coronavirus: host-pathogen interaction

Filling the void: proximity-based labeling of proteins in living cells

Efficient proximity labeling in living cells and organisms with TurboID

A promiscuous biotin ligase fusion protein identifies proximal and interacting proteins in mammalian cells

Spatially resolved proteomic mapping in living cells with the engineered peroxidase APEX2

An improved smaller biotin ligase for BioID proximity labeling

Determination of host proteins composing the microenvironment of coronavirus replicase complexes by proximity-labeling

Directed evolution of APEX2 for electron microscopy and proximity labeling

Electron microscopy using the genetically encoded APEX2 tag in cultured mammalian cells

Atlas of subcellular RNA localization revealed by APEX-Seq

BioID: a screen for protein-protein interactions

Processing of the coronavirus MHV-JHM polymerase polyprotein: identification of precursors and proteolytic products spanning 400 kilodaltons of ORF1a

RNA replication of mouse hepatitis virus takes place at double-membrane vesicles

Coronavirus replicase-reporter fusions provide quantitative analysis of replication and replication complex formation

Recombinant mouse hepatitis virus strain A59 from cloned, full-length cDNA replicates to high titers in vitro and is fully pathogenic in vivo

Generation of recombinant coronaviruses using vaccinia virus as the cloning vector and stable cell lines containing coronaviral replicon RNAs

Sequential use of milk and bovine serum albumin for streptavidin-probed western blot

A pneumonia outbreak associated with a new coronavirus of probable bat origin