Identification of the interacting partners of a lysosomal membrane protein in living cells by BioID technique

Summary The purpose of this protocol is to screen and identify the physiologically relevant interactors of a lysosomal protein in living cells. Here, we describe how to identify solute carrier family 15 member 4 (SLC15A4)-interacting proteins by BioID and mass spectrometry analysis. This protocol utilizes fusion of SLC15A4 with a mutant form of biotin ligase, BirA. The fusion protein can promiscuously biotinylate the proteins proximal to SLC15A4. The biotinylated endogenous proteins are pulled down by magnetic streptavidin beads and detected by mass spectrometry analysis. For complete details on the use and execution of this protocol, please refer to Kobayashi et al. (2021).


Identification of the interacting partners of a lysosomal membrane protein in living cells by BioID technique
The purpose of this protocol is to screen and identify the physiologically relevant interactors of a lysosomal protein in living cells. Here, we describe how to identify solute carrier family 15 member 4 (SLC15A4)-interacting proteins by BioID and mass spectrometry analysis. This protocol utilizes fusion of SLC15A4 with a mutant form of biotin ligase, BirA. The fusion protein can promiscuously biotinylate the proteins proximal to SLC15A4. The biotinylated endogenous proteins are pulled down by magnetic streptavidin beads and detected by mass spectrometry analysis.

SUMMARY
The purpose of this protocol is to screen and identify the physiologically relevant interactors of a lysosomal protein in living cells. Here, we describe how to identify solute carrier family 15 member 4 (SLC15A4)-interacting proteins by BioID and mass spectrometry analysis. This protocol utilizes fusion of SLC15A4 with a mutant form of biotin ligase, BirA. The fusion protein can promiscuously biotinylate the proteins proximal to SLC15A4. The biotinylated endogenous proteins are pulled down by magnetic streptavidin beads and detected by mass spectrometry analysis. For complete details on the use and execution of this protocol, please refer to Kobayashi et al. (2021).

BEFORE YOU BEGIN
This protocol was used to identify the interacting proteins of solute carrier family 15 member 4 (SLC15A4), a lysosome-resident amino acid/oligopeptide transporter, by BioID and mass spectrometry analysis (Kobayashi et al., 2021). The co-immunoprecipitation (Co-IP) is a conventional technique to identify protein-protein interactions. However, it has limitations in membrane proteins like SLC15A4 because of protein solubility or aggregation/filament formation in Co-IP lysis buffer. BioID, a new method to screen for physiologically relevant protein interactions based on a promiscuous biotin ligase (BirA) in a natural cellular setting (Roux et al., 2013), has advantages to overcome the Co-IP limitations because all of the cellular proteins are denatured and solubilized, and only those that are labeled with biotin are selectively captured. Here, we generated Control-BirA or BirA-SLC15A4-expressing HEK293T cells. The fusion ligases biotinylate adjacent endogenous proteins, which can be pulled down by magnetic streptavidin beads. These biotinylated proteins were identified by mass spectrometry. After identifying biotinylated proteins in each sample, we compared the proteins detected in the control samples (BirA only) with the BirA-SLC15A4 samples. The proteins enriched in the BirA-SLC15A4 sample represent the potentially SLC15A4-interacting candidates. and further subcloned into pEBMulti-Bsd (FUJIFILM Wako Chemicals) to obtain the blasticidin S-resistance. BirA protein was fused at the N-terminal of SLC15A4. BirA-SLC15A4 expression vector was transfected into HEK293T cells using Lipofectamine 3000 Transfection Reagent (Thermo Fisher Scientific) to generate stably expressing cells; stable transfectants were selected in 10 mg/mL blasticidin S (FUJIFILM Wako Chemicals). We created a control cell line by transfecting HEK293T cells with a pEBMulti-Met-Gly-Cys-cMyc-BirA vector (Cont-BirA). The MGC motif helps to anchor cMyc-BirA to the cell membrane, which is related subcellular localization with BirA-SLC15A4.

Timing: 3-4 days
We performed flow cytometric analysis with c-Myc staining to confirm the expression of BioID constructs and western blotting with streptavidin-HRP (Thermo Fisher Scientific) to confirm the activity of BioID before scaling up for BioID pull down assays. If necessary, single cell cloning by limited dilution should be performed to choose the cell lines with comparable expression level. If necessary, cell viability and/or cytotoxicity assays should be performed because some recombinant fusion proteins have cytotoxic effects.
In addition, it is important to confirm the subcellular localization of the BirA fusion protein using fluorescence imaging whether it shows the same localization as the endogenous protein. In our case, we confirmed that both endogenous SLC15A4 and BirA-SLC15A4 showed the localization in LAMP1 + compartment.  Preparation of samples with sufficient amount of protein biotinylated by the BioID fusion protein is identified by mass spectrometry (Figure 1). At the starting point, the amount of total protein per sample is about 20 mg.

KEY RESOURCES
Note: Prepare biological triplicate samples to see reproducibility in the next mass spectrometric analysis. All steps hereafter are carried out using DNase/RNase-free tubes that have not previously opened and a tube cap opener, and wearing gloves, face mask, safety goggle and head net, to avoid contamination of the samples, particularly with keratin. Should perform the experiment inside a clean bench. a. Remove medium completely by aspiration. b. Collect the cells by pipetting with 10 mL/dish of PBS/EDTA into 15 mL tube. c. Wash the dish one more time with 5 mL PBS/EDTA to collect all residue cells. d. Spin down for 5 min at 400 3 g at room temperature (20 C-25 C). 5. Discard the supernatant and wash the cells twice by spin down for 5 min at 400 3 g at room temperature (20 C-25 C) with 10 mL/tube of PBS to remove all residue biotin. 6. Lyse the cells. This step is to completely disrupt all protein interactions and denature proteins.
a. Resuspend the cells in 900 mL cell-suspension buffer (the cells need to be completely suspended for sufficient lysis). b. Add 900 mL lysis buffer and pipette strongly the solution but avoid making the bubble. Perform this step at room temperature (20 C-25 C).
Note: If it is hard to pipette the solution due to DNA genome releasing, apply a short sonication at 30% amplitude for 15-30 s at room temperature (20 C-25 C).
7. Add 180 mL of 20% Triton X-100 and mix by pipetting (to prevent SDS precipitation on ice). From now keep the tubes on ice for subsequent sonication. 8. Position the sonicator probe tip in the sample just above the tube bottom to avoid making the bubble. 9. Sonicate samples.
a. Apply sonication for two sessions for 30 s using an ultrasonic homogenizer at 30% amplitude. b. Then let the tube rest on ice for 2 min between each session to prevent overheating. c. During the resting time, apply the sonication for other samples after washing the probe tip with 70% ethanol and ultra-pure water. 10. Add 1.62 mL of prechilled 50 mM Tris-HCl, pH 7.5, and mix well. 11. Repeat one more session of sonication as mentioned in steps 8 and 9. 12. Aliquot the sample evenly into three prechilled 1.5-mL tubes.
Note: This step can be excluded if the rotor for 5 ml tube is available.
13. Spin down for 10 min at 16,500 3 g, 4 C. Carefully transfer all supernatant of the same sample to a prechilled 5 mL tube.
Note: Do not disturb the small insoluble pellet on the tube wall when transferring the supernatant.
14. Measure protein concentration by using the DC protein assay kit (microplate assay) with BSA as a standard. Adjust the protein concentration of samples to the lowest concentration samples with the mix buffer. Keep samples on ice until use.
Perform affinity purification of biotinylated proteins (Steps 15-17 are to equilibrate the magnetic streptavidin beads with the mix buffer).
15. During steps 13 and 14, place three new 1.5-mL tubes in the Tamagawa magnetic stand (TA4899N12). Add 1 mL of the mix buffer to each tube. 16. Mix the stock of magnetic streptavidin beads well with gentle tapping to resuspend and add 150 mL of beads to each tube prepared in step 15. Wait for 3 min at room temperature (20 C-25 C). 17. Once the beads accumulate at one side of tube wall, remove the supernatant gently by 1-mL pipette.
(Continue with the samples from step 14).
18. Divide 1.2 mL of sample from step 14 to the three tubes containing equilibrated beads. 19. Resuspend the samples and beads with gentle pipetting. 20. Incubate the tube on a rotator at 4 C for overnight (at least 12 h). 21. Place the tubes on the magnetic stand and wait for 3 min to collect beads at room temperature (20 C-25 C). 22. Remove the supernatant gently by pipetting. Try not to disturb beads on the wall of the tube. 23. Add 1.2 mL of wash buffer 1 to each tube and resuspend beads gently by pipetting. 24. Rinse the beads with rotation in a rotator for 8 min at room temperature (20 C-25 C). 25. Remove the supernatant as described in steps 21 and 22.
(Pool beads from same samples into one tube).
26. Add 1.2 mL of wash buffer 1 to one of the three tubes and resuspend gently. 27. Transfer resuspended sample from step 26 to the second of the three tubes, and resuspend gently. Repeat this step to pool all beads from same samples into one tube in a final volume of 1.2 mL wash buffer 1.
Note: From this step, it is easy to lose beads, be careful with pipetting ( Figure 2).
28. Place the combined tube on a rotator for 8 min at room temperature (20 C-25 C). 29. Wash once with 1.2 mL of wash buffer 2 as described in steps 21 through 25.

OPEN ACCESS
30. Wash once with 1.2 mL of wash buffer 3 as described in steps 21 through 25. 31. Wash once with 1.2 mL of 50 mM Tris-HCl, pH 7.4 as described in steps 21 through 25. (troubleshooting 1). 32. Remove the supernatant completely. Do not disrupt beads at the tube wall. 33. Add 110 mL Laemmli SDS sample buffer containing 2 mM biotin. 34. Save 10 mL (10% of total) of sample for further analysis by silver stain or western blot (Figure 3). 100 mL of sample will be used for mass spectrometric analysis.  Pause Point: If necessary, freeze the sample quickly in liquid nitrogen. Frozen samples can be stored at À80 C until performing mass spectrometric analysis.

Timing: 2 days
Single-band focused SDS-PAGE containing whole proteins and its in-gel digestion with trypsin ( Figure 4).
Note: Prepare biological triplicate samples to see reproducibility in the next mass spectrometric analysis. To avoid modification by unreacted acrylamide, the one-day-old gel is used. All steps after are carried out not to contaminate the samples with keratin.
35. Elute the streptavidin bound proteins with 100 mL Laemmli SDS sample buffer containing 2 mM biotin at 98 C. 36. Concentrate the samples to the volume of 20 mL with a SpeedVac vacuum concentrator for 2 h at 25 C. 37. Load the concentrated samples onto a polyacrylamide mini gel.
Note: If there are multiple samples, load each sample onto a gel at a distance to avoid crosscontamination.
38. Stop the electrophoresis when the dye front migrated from the stacking gel to the separation gel. 39. Wash the gel three times with 200 mL of ultra-pure water using a shaker and discard the washing solution. 40. Stain the gel according to the CGP stain method (Yasumitsu et al., 2010). Pour 50 mL of the CGP staining solution into the gel, and incubate for at least 1 h at 25 C. 41. Discard the staining solution and wash the gel three times with 200 mL ultra-pure water for 1 h to visualize the protein bands. In each lane, a single band is observed as a mixture of all the proteins. 42. Cut the visualized band into the four pieces using a clean razor blade and transfer them to a 600 mL microcentrifuge tube. 43. Soak the gel pieces in 300 mL of de-staining buffer to de-stain CBB G-250. Incubate for 30 min at 37 C and exchange the de-staining buffer until the gel pieces turn clear. 44. Dehydrate the gel pieces by adding 100% acetonitrile. Exchange 100% acetonitrile until the gel pieces turn white. Discard the acetonitrile and dry the gel pieces completely with a SpeedVac vacuum concentrator for 10 min at 25 C. 45. Add 50 mL of reducing buffer onto the dried gel pieces and incubate for 2 h at 37 C. 46. Add 50 mL of alkylating buffer onto the reduced gel pieces and incubate for 30 min at 25 C. Remove the buffer after alkylation by pipetting. 47. Wash gel pieces three times with 200 mL of ultra-pure water for 10 min at 25 C. Exchange washing solution with 200 mL of 100% acetonitrile and reswell with 200 mL of ultra-pure water. This step is repeated three times.
Note: It is advisable to wash over three times to remove any reagents that may affect the subsequent mass spectrometric analysis.
48. Remove the washing solution and dehydrate the gel pieces with 100% acetonitrile, Dry the gel pieces with a centrifugal vacuum concentrator for 10 min at 25 C. 49. Reswell the dried gels completely with 2 mL of 10 ng/mL TPCK-treated trypsin. Add 30 mL of digestion buffer to cover the top of gel pieces.

OPEN ACCESS
Note: The standard protocol is for about 1 mg protein, and the amount of enzyme and buffer are changed according to the amount of protein.
50. Incubate samples for 12 h at 37 C to digest with trypsin. The resulting peptides can be stored at 4 C for 1 week. Pause Point: At this time, the peptide mixture can be safely stored at À20 C for one month.

LC-MS/MS analysis for label-free proteomics of BioID pull-down proteins
Timing: 1 day Peptide mixture generated by carrying out steps 35-50 is analyzed using LC-MS/MS. The obtained data are used for the next label-free quantitative analysis.
Note: Prepare biological triplicate samples to see reproducibility and all samples are analyzed in the same volume for use in the next step of comparative quantification.
51. Transfer the digested peptides mixture to the vials in the LC autosampler and acidify the peptides solution by adding 1% formic acid. Adjust all sample volumes to equal. (troubleshooting 2). 52. Analyze the digestion mixtures with a nanospray column Nikkyo Technos,Co.,Ltd.) using EASY-nLC 1200 (Thermo Fisher Scientific) coupled with a Q Exactive HF-X mass spectrometer (Thermo Fisher Scientific). LC-MS/MS data of three independent biological replicates are available through the ProteomeXchange: PXD020370) (troubleshooting 3). a. A flow rate of 300 nL/min was used and peptides were separated using a 50 min linear gradient of 0%-64% over B for 50 min (solvent A: 0.1% formic acid in water, solvent B: 0.1% formic acid in 80% acetonitrile). b. Carry out LC-MS/MS analysis according to the following parameters in Table 1. LC-MS/MS parameters are listed in Table 1.

EXPECTED OUTCOMES
To check our protocol works well, an intensity correlation between BioID experimental samples and control samples was plotted in Figure 5. Pyruvate carboxylase (PC), an endogenous biotin-binding protein, and SLC15A4, a BirA fusion protein, are shown on the plot. SLC15A4 was detected with the highest intensity only in the BioID experimental sample, while PC was strongly detected in both BioID experimental samples and control samples. The results showed that BirA-SLC15A4 was specifically and strongly biotinylated by BioID experiments. Also, proteins with a higher abundance ratio than the endogenous PC are likely to have been specifically biotinylated by BioID experiments. Therefore, the proteins in the region between the dashed lines were treated as candidates for biotinylated proteins in the BioID experiment. As a result, 629 proteins could be identified as biotinylated proteins by BioID experiments. Furthermore, as indicated in the section on quantitative and statistical analysis and troubleshooting, it is possible to verify the validity of the results by creating several constructs and conducting the same experiment. Among the identified proteins, we could confirm the interaction between SLC15A4 and the lysosomal proteins involved in mTOR pathway by Co-IP assay (Kobayashi et al. 2021).

QUANTIFICATION AND STATISTICAL ANALYSIS
Timing: 1 day The LC-MS/MS data is processed by label-free proteomic analysis and the list of proteins that have been specifically increased by the BioID experiment is obtained. (troubleshooting 4, 5, 6).
Note: The use of data from biological triplicate samples increases the reliability of the results.
1. Analyze LC-MS/MS data using Proteome Discoverer 2.2 (Thermo Scientific) combined with Mascot 2.6 (Matrix Science) for protein identification and label-free quantification.
Note: The analysis software should be used Proteome Discoverer version 2.2 or higher. Also, Other MS analysis software like SEQUEST (Yates et al. 1995) and MaxQuant (Tyanova et al., 2016) can be used.

Problem 2
Difficult to de-stain CGP stain (step-by-step method details step 51).

Potential solution
Proceed to the alkylation step and then carry out the de-staining step.

Problem 3
Low detection sensitivity of peptides in LC-MS (step-by-step method details step 52).

Potential solution
If the protein amounts are too low, start by re-preparing BioID samples to increase amounts of samples.

Potential solution
Confirmation of in-gel digestion steps by peptide mass fingerprinting using MALDI-TOF MS.

Potential solution
If it is an LC-MS problem, use a standard sample (e.g., tryptic Hela cell lysate) to check the condition of the instrument. For example, we analyzed 0.5 mg of a proteome sample of HL60 cells using our instruments and settings. As a result, 940 proteins and 2,967 peptides were identified with high confidence.

Problem 4
Problems of quantification accuracy due to the accuracy of gel excision (quantification and statistical analysis).

Potential solution
If further accuracy is required, digestion in a solution system should be considered.

Problem 5
Differences in the efficiency of in-gel digestion between samples (quantification and statistical analysis).

Potential solution
Examine the number of missed cleavages per sample and monitor the variability.

Problem 6
Many false-positive candidates (quantification and statistical analysis).

Potential solution
Perform BioID experiments with other proteins like the target protein, or proteins with BirA fused to the N-and C-termini respectively. The candidate proteins can be validated by comparing proteins that show similar behavior in these multiple experiments.

RESOURCE AVAILABILITY
Lead contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Noriko Toyama-Sorimachi (nsorimachi@ri.ncgm.go.jp).

Materials availability
All reagents generated in this study are available from the lead contacts upon completing a Materials Transfer Agreement. Data and code availability LC-MS/MS raw files and Proteome Discoverer output files are available at ProteomeXchange: PXD020370.