Protocol to purify and detect ubiquitinated phospholipids in budding yeast and human cell lines

Summary Ubiquitin is covalently conjugated to phospholipids as well as proteins; however, ubiquitinated phospholipids are less abundant than free ubiquitin and ubiquitinated proteins. Here, we describe protocols to purify ubiquitinated phospholipids in budding yeast and human cells based on their hydrophobicity. Ubiquitinated phospholipids are purified by Triton X-114 phase partitioning and affinity purification and verified by phospholipase D treatment. These protocols enable the detection of tagged as well as endogenous mono- and poly-ubiquitinated phospholipids by immunoblotting. For complete details on the use and execution of this protocol, please refer to Sakamaki et al..1


SUMMARY
Ubiquitin is covalently conjugated to phospholipids as well as proteins; however, ubiquitinated phospholipids are less abundant than free ubiquitin and ubiquitinated proteins. Here, we describe protocols to purify ubiquitinated phospholipids in budding yeast and human cells based on their hydrophobicity. Ubiquitinated phospholipids are purified by Triton X-114 phase partitioning and affinity purification and verified by phospholipase D treatment. These protocols enable the detection of tagged as well as endogenous mono-and poly-ubiquitinated phospholipids by immunoblotting. For complete details on the use and execution of this protocol, please refer to Sakamaki et al.. 1

BEFORE YOU BEGIN
Ubiquitination is a common post-translational protein modification that regulates the stability, localization, activity, and complex formation of target proteins. [2][3][4] We have recently shown that the phospholipid phosphatidylethanolamine (PE) is also ubiquitinated in budding yeast and human cells. 1 However, because ubiquitinated PE (Ub-PE) is less abundant than free ubiquitin and ubiquitinated proteins, it is difficult to detect Ub-PE by immunoblotting of whole cell lysates. Thus, we describe protocols for the efficient purification of Ub-PE from yeast and human cells. When conjugated to PE, ubiquitin becomes hydrophobic and can be separated by Triton X-114 phase partitioning. [5][6][7][8] Triton X-114 has a low cloud point of approximately 23 C, and a buffer containing Triton X-114 separates into the detergent and aqueous phases above the cloud point. Hydrophobic proteins, including lipid-conjugated proteins and membrane proteins, are incorporated into the detergent phase owing to their hydrophobicity. On the other hand, free ubiquitin and soluble ubiquitinated proteins, which are abundant in cells, are incorporated into the aqueous fraction and can be removed. Hydrophobic ubiquitin species in the detergent fraction are further purified by affinity purification (Figure 1). Ub-PE is detected at a slightly higher position than that of the unmodified monomeric ubiquitin on an immunoblot. Phospholipase D, which hydrolyzes a phosphodiester bond of glycerophospholipids, causes a downward shift of Ub-PE, validating the conjugation of ubiquitin to PE. Below are the procedures to purify endogenous and tagged Ub-PE from yeast and human cells. b. Collect log-phase cells (OD 600 z 1.0) in a 15-mL tube by centrifugation at 1,000 3 g for 5 min. c. Wash cells with 1 mL of ddH 2 O and proceed to step 1 of the step-by-step method details. 2. To detect endogenous Ub-PE, collect cells as described below.

Yeast cell culture
a. Inoculate 50 mL of yeast extract-peptone-dextrose (YPD) medium in a 300-mL flask with BY4741 cells and culture with shaking at 150 rpm at 30 C.
CRITICAL: 2a should be performed under sterile conditions.
b. Collect log-phase cells (OD 600 z 1.0) in a 50-mL tube by centrifugation at 1,000 3 g for 5 min. c. Wash cells with 5 mL of ddH 2 O and proceed to step 1 of the step-by-step method details.
a. Seed HeLa cells stably expressing 3xFLAG-Ub in DMEM complete medium into two 10-cm dishes.
CRITICAL: 3a-b should be performed under sterile conditions. b. Culture cells for 2-3 days until cells reach 80% confluence. c. Remove the medium and wash cells with 10 mL of ice-cold phosphate-buffered saline (PBS). d. Collect the cells by scraping in 1 mL of ice-cold PBS and transfer them to a 1.5-mL tube. e. Centrifuge at 1,000 3 g at 4 C for 5 min. f. Remove the supernatant and proceed to step 2 of the step-by-step method details. 4. To detect endogenous Ub-PE, collect cells as described below.
a. Seed HeLa cells in DMEM complete medium into ten 10-cm dishes.
CRITICAL: 4a-c should be performed under sterile conditions. Total membranes are isolated from yeast or HeLa cells expressing 33FLAG-Ub and solubilized in a buffer containing 1% Triton X-114. The sample is subjected to Triton X-114 phase partitioning followed by immunoprecipitation of 33FLAG-Ub in the Triton X-114 fraction. Hydrophobic ubiquitin species contain ubiquitinated phospholipids and hydrophobic proteins (e.g., membrane proteins).

OPEN ACCESS
b. Culture cells for 2-3 days until cells reach 80% confluence. c. Remove the medium and wash cells with 10 mL of PBS. d. Add 1 mL of ice-cold PBS to each dish, collect the cells by scraping, and transfer them to a 15-mL tube. e. Centrifuge at 1,000 3 g at 4 C for 5 min. f. Remove the supernatant and proceed to step 2 of the step-by-step method details. h. Remove cell debris by centrifugation at 4 C at 17,700 3 g for 10 min. i. Pass the supernatant through a 0.45-mm pore filter and transfer it to a 15-mL tube. j. Incubate the supernatant with 1 mL of glutathione Sepharose 4B (50% in CelLytic B buffer) at 4 C for 12-18 h using a rotator. k. Wash the beads with 1 mL of ice-cold CelLytic B reagent three times. l. Wash the beads with 1 mL of ice-cold Triton X-114 wash/dilution buffer three times. m. Store the beads in 500 mL of Triton X-114 wash/dilution buffer (50% in buffer) at 4 C. n. To confirm that GST-UQ1 UBA is properly generated, mix 5 mL of GST-UQ1 UBA with 1 mL of 63 SDS-PAGE sample buffer and perform SDS polyacrylamide gel electrophoresis (SDS-PAGE) followed by Coomassie brilliant blue (CBB) staining. GST-UQ1 UBA should be detected at approximately 31 kDa.
Note: To spin down the beads, centrifuge at 4 C at 1,000 3 g for 1 min.

MATERIALS AND EQUIPMENT
CRITICAL: Glucose should be separately prepared as a solution and autoclaved. Glucose solution should be added to the medium under sterile conditions.
Dissolve yeast extract and bacto peptone in 900 mL of ddH2O and autoclave the mixture.
Dissolve 20 g of glucose in 100 mL of ddH2O and autoclave the solution.
Add 100 mL of autoclaved 20% glucose to 900 mL of the autoclaved yeast extract and bacto peptone mixture.
Store at 22 C-25 C. Stable for at least a year. CRITICAL: Glucose, adenine, and L-tryptophan should be separately prepared as solutions and sterilized. Glucose solution and the supplements should be added to the medium under sterile conditions.
Dissolve yeast nitrogen base, ammonium sulfate, and casamino acids in 880 mL of ddH2O, add 750 mL of 5 N NaOH, and autoclave the mixture.
Dissolve 20 g of glucose in 100 mL of ddH2O and autoclave the solution.
Dissolve 200 mg of adenine in 100 mL of ddH2O and sterilize the solution using a 0.22-mm pore filter.
Dissolve 200 mg of l-tryptophan in 100 mL of ddH2O and sterilize the solution using a 0.22-mm pore filter.
Add 100 mL of autoclaved 20% glucose, 10 mL of 2 mg/mL adenine, and 10 mL of 2 mg/mL l-tryptophan to 880 mL of the autoclaved yeast nitrogen base, ammonium sulfate, and casamino acids mixture.
Store at 22 C-25 C. Stable for at least a year.

Reagent Final concentration Amount
Yeast nitrogen base without amino acids and ammonium sulfate 0. Store at À20 C. Stable for at least a year.

Timing: 2 h
This section describes the isolation of total membranes from 12-mL cultures of budding yeast and two 10-cm dishes of HeLa cells. To detect endogenous Ub-PE in 50 mL of yeast culture or ten 10-cm dishes of HeLa cells, use 53 volumes of buffers throughout the procedures described below.
1. Isolation of total membrane from yeast cells. a. Resuspend cells in 500 mL of ice-cold disruption buffer containing protease inhibitor cocktail. b. Transfer the cell suspension to a 2-mL tube (e.g., Watson Bio Lab, Cat# 1392-200) containing approximately 500 mL of 0.5 mm zirconium dioxide beads. c. Disrupt cells using a multi-beads shocker (Yasui Kikai, 2700 rpm, 30 s on/30 s off, twice). d. Remove intact cells, cell debris, and nuclei by centrifugation at 1,000 3 g at 4 C for 5 min. e. Collected 25 mL of the supernatant and mix it with 5 mL of 63 sample buffer (total lysate). f. Apply the supernatant to ultracentrifugation at 100,000 3 g at 4 C for 1 h. troubleshooting 3. g. Remove the supernatant (the cytosol) and proceed to step 3.
Note: To efficiently recover the lysate at step d, puncture the bottom of the 2-mL tube using an 18-gauge needle, place it in a 15-mL tube, and carry out centrifugation.

Isolation of total membrane from HeLa cells.
a. Resuspend cells in 500 mL of ice-cold homogenization buffer. b. Centrifuge at 4 C at 1,000 3 g for 5 min. c. Remove the supernatant, resuspend cells in 500 mL of ice-cold homogenization buffer containing protease inhibitor cocktail, and incubate on ice for 10 min. d. Disrupt cells using a 27-gauge needle and syringe unit (20 strokes). e. Centrifuge at 4 C at 1,000 3 g for 5 min to remove intact cells, cell debris, and nuclei.  f. Collect 25 mL of the supernatant and mix it with 5 mL of 63 sample buffer (total lysate). g. Apply the supernatant to ultracentrifugation at 100,000 3 g at 4 C for 1 h. troubleshooting 3. h. Remove the supernatant (i.e., the cytosol) and proceed to step 3.
Note: Cells can also be disrupted by nitrogen decompression (Parr Instrument Company, model# 4639 Cell Disruption Vessel, 500 psi).

Purification of hydrophobic proteins
Timing: 2 h This section describes the purification of hydrophobic proteins by Triton X-114 phase partitioning. When performing the delipidation assay for verification of conjugation of ubiquitin to phospholipids, proceed to step 7 after completing this step (see also step 7 for details).
3. Purification and concentration of hydrophobic ubiquitin. a. Resuspend the pellet (i.e., membranes) in 200 mL of ice-cold Triton X-114 buffer containing protease inhibitor cocktail. b. Centrifuge at 4 C at 16,900 3 g for 10 min to remove the insoluble fraction. c. Transfer the supernatant to a new 1.5-mL tube. d. Place the tube at 37 C for approximately 1 min (until the solution becomes cloudy). e. Centrifuge at 25 C at 16,900 3 g for 10 min. f. Remove the top layer (i.e., the aqueous phase, approximately 180 mL in the first round and 500 mL in the wash steps). g. Add 500 mL of Triton X-114 wash/dilution buffer and mix well. h. Place the tube on ice until the solution becomes clear (within several min). i. Repeat steps 3d-3h twice (for a total of three repetitions of wash steps). troubleshooting 4. j. After the third repetition of wash steps, add 500 mL of Triton X-114 wash/dilution buffer containing a proteinase inhibitor cocktail to the detergent phase. k. Place the tube on ice until the solution becomes clear (within several min). l. Centrifuge at 4 C at 16,900 3 g for 10 min to remove debris. m. Transfer approximately 520 mL of the supernatant to a new 1.5-mL tube. n. Proceed to step 4 or 5.
Note: Buffers containing Triton X-114 separate into the detergent and aqueous phases above the cloud point (approximately 23 C) and become cloudy.

Timing: 1 day
This section describes the purification of hydrophobic ubiquitin by immunoprecipitation with anti-FLAG antibody (step 4) and pull-down with ubiquilin 1 ubiquitin-associated domain (UQ1 UBA) 9 (step 5).

Immunoprecipitation of 3xFLAG-Ub with anti-FLAG antibody
a. Wash anti-FLAG M2 affinity beads with 1 mL of Triton X-114 wash/dilution buffer three times. b. Add 20 mL of anti-FLAG M2 affinity beads (50% in Triton X-114 wash/dilution) to the sample. c. Incubate at 4 C for 12-18 h using a rotator. d. Wash the beads with 500 mL of Triton X-114 wash/dilution buffer three times. e. Wash the beads with 500 mL of ddH 2 O twice. f. Proceed to step 6 or add 50 mL of 13 sample buffer and proceed to step 8.

OPEN ACCESS
Note: To spin down the beads, centrifuge at 4 C at 1,000 3 g for 1 min.
Pause point: After mixing with 13 sample buffer, the samples can be stored at À20 C until until proceeding to SDS-PAGE and immunoblotting.
5. Purification of endogenous ubiquitin using GST-UQ1 UBA a. Add 20 mL of GST-UQ1 UBA immobilized on glutathione Sepharose (50% in Triton X-114 wash/dilution buffer). b. Incubate at 4 C for 12-18 h using a rotator. c. Wash the beads with 500 mL of Triton X-114 wash/dilution buffer three times. d. Wash the beads with 500 mL of ddH 2 O twice. e. Proceed to step 6 or add 50 mL of 13 sample buffer and proceed to step 8.
Pause point: After mixing with 13 sample buffer, the samples can be stored at À20 C until until proceeding to SDS-PAGE and immunoblotting.

Timing: 1.5-2.5 h
To verify that ubiquitin is conjugated to phospholipids, the sensitivity of hydrophobic ubiquitin to phospholipase D (PLD) can be tested. This section describes on-bead PLD treatment and a delipidation assay using the Triton X-114 fraction in step 3. In the delipidation assay, the purified Triton X-114 fraction is incubated with PLD and subjected to another round of phase partitioning (Figure 2). After PLD treatment, delipidated ubiquitin species are remobilized from the Triton X-114 to the aqueous fraction, while hydrophobic proteins (e.g., ubiquitinated membrane proteins) remain in the Triton X-114 fraction.  iv. Develop using Immobilon western chemiluminescent horseradish peroxidase substrate. v. Acquire images using a Fusion Solo 7S system (Vilber).

EXPECTED OUTCOMES
Higher molecular weight ubiquitin monomers are detected in the Triton X-114 fraction. This band moves downwards upon phospholipase D treatment (Figures 3A-3C).
In the delipidation assay, mono-and poly-ubiquitinated ubiquitin species move from the Triton X-114 fraction to the aqueous phase upon phospholipase D treatment ( Figure 3D).

LIMITATIONS
A large amount of poly-ubiquitinated species is detected in the Triton X-114 fraction. However, it is difficult to assess whether these species are poly-ubiquitinated PE, because ubiquitinated membrane proteins are also enriched in the Triton X-114 fraction. To detect poly-ubiquitinated PE, perform the delipidation assay as described in step 7.
Ub-PE is more sensitive to antibodies compared to unmodified ubiquitin as also reported for ATG8. [11][12][13] It is therefore difficult to accurately estimate the percentage of Ub-PE by this method.

TROUBLESHOOTING Problem 1
Weak Ub-PE signal (yeast and human cell culture in before you begin).

Potential solution
Although Ub-PE is detected under basal conditions, its level is not high. To address this, increase the cell number or use a stimulus, such as nitrogen starvation (3 h) and concanamycin A (500 nM, 3 h) (yeast cells) or Torin 1 (500 nM, 24 h) and bafilomycin A 1 (100 nM, 6 h) (human cells), to increase Ub-PE levels.

Problem 2
Low efficiency in ubiquitin pull-down (preparation of GST-UQ1 UBA in before you begin).

Potential solution
Check that GST-UQ1 UBA is properly produced by conducting SDS-PAGE and CBB staining. The typical yield is approximately 1-2 mg/mL.

Problem 3
Weak or no Ub-PE signal (isolation of total membrane from yeast and HeLa cells in steps 1 and 2).

Potential solution
Check that membranes are retrieved after ultracentrifugation. To do this, add 50 mL of 13 sample buffer to the pellet and perform immunoblotting. It is important to confirm that endosomal and vacuolar (or lysosomal) markers are detected in the membrane fraction, because ubiquitinated PE is mainly localized to these organelles. 1 The following antibodies can be used:

Problem 4
Contamination of soluble proteins (purification of hydrophobic proteins in step 3).

Potential solution
Increase the repetitions of the wash steps.

Problem 5
No clear molecular weight shift of Ub-PE (SDS-PAGE in step 8).

Potential solution
Use 15% gel and run the samples longer for better separation among low molecular weight species.

RESOURCE AVAILABILITY
Lead contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Noboru Mizushima (nmizu@m.u-tokyo.ac.jp).

Materials availability
This study did not generate new unique reagents.

Data and code availability
This study did not generate/analyze datasets/code.