From 1e6 cells (Hela) you can get about 100 ug of protein

Cell Culture (from Ref 4, chapter 10)

Materials

1. Eight culture flasks (175 cm 2 growth area) of confluent HEK293T cells.
2. Versene (EDTA-saline).
3. Phosphate buffered saline (PBS), without calcium or magnesium, pH 7.2.

1. Decant culture medium from cell culture flasks and wash cells with PBS. Decant PBS and add 4 mL Versene to each 175 cm 2 culture flask. Incubate at 37 ^oC for 5 min to allow Versene to disrupt cell adhesion ( The effectiveness of Versene for detaching cells is dependent on the adhesion proteins expressed by a particular cell line. For example, while Versene is very effective at detaching fibroblasts such as HEK293T, E14TG2a murine embryonic stem cells cultured on gelatinized flasks are not effectively dissociated. Other enzyme and detergent-free dissociation buffers may be suitable alternatives to Versene for such cell lines. Cells may also be detached by rapid trypsin treatment. To minimize plasma membrane protein degradation, care should be taken to quench trypsin with an excess of a serum-containing medium as soon as cells are detached, after which the cells should be placed on ice immediately. The trypsin is then diluted out in subsequent PBS washes, and the cells are resuspended in a lysis buffer containing protease inhibitors that will nullify residual trypsin activity). Detach cells from the flask by gentle tapping. Transfer detached cells into two 50 mL tubes, dilute to tube capacity with PBS, and place on ice.
2. Pellet cells by centrifugation at 200 x g for 5 min at 4 ^oC. Decant the supernatant and gently resuspend the cell pellets in a total volume of 50 mL PBS. Transfer the cell suspension into a single 50 mL tube, and centrifuge as before to re-pellet the cells. Repeat this wash procedure once more.
3. Resuspend the cell pellet in 15 mL detergent-free lysis buffer and place on ice for 5 min.
4. Set up the ball bearing homogenizer on ice; place the appropriate tungsten carbide ball bearing into the homogenizer chamber so that the clearance diameter is 12 μm (The clearance space is the diameter between the tungsten carbide ball bearing and the sides of the stainless steel homogenizer chamber. This can be attenuated by using ball bearings of different sizes, and should be optimized for a particular cell line or application. The aim is to set the clearance space so that it is too narrow for cells to pass through intact, but permissive enough to allow organelles such as nuclei and mitochondria to pass through with minimal disruption. Narrower clearances will increase the back pressure required for homogenization, while wider clearances may reduce lysis efficiency. Ball bearing homogenizers are highly suited to this application because of their consistent performance and high lysis efficiency. If a ball bearing homogenizer is unavailable, then alternative mechanical homogenization methods such as Dounce homogenization may be suitable), and displace air from the homogenizer chamber by passing 1 mL detergent-free lysis buffer back and forth between the two 1 mL syringes.
5. Gently invert the cell suspension several times to resuspend any settled material. Draw 1 mL aliquots of the solution into a 1 mL syringe and pass the solution through the ball bearing homogenizer and between two syringes 15 times (The optimal number of passages through the ball bearing homogenizer may require some optimization for each cell line).
6. To reduce sample viscosity, add 500 U Benzonase to the cell lysate and incubate at room temperature for 20 min, with occasional mixing by inversion. Place the lysate back on ice for an additional 10 min to cool. The lysate may be stored at 4 ^oC overnight at this point if necessary. The lysate cannot be frozen, as freeze-thawing disrupts organelle integrity.

Cell Lysis and fractionation (from Ref 4, chapter 10)

Materials

1. 1. 0.1 M HEPES-HCl (pH 7.4), store at 4 ^oC.
2. 0.1 M EDTA (pH 8), store at 4 ^oC.
3. 0.1 M magnesium acetate tetrahydrate, store at 4 ^oC.
4. Detergent-free lysis buffer (50 mL per experiment): 0.25 M sucrose, 10 mM HEPES-HCl (pH 7.4), 2 mM EDTA, 2 mM magnesium acetate tetrahydrate, protease inhibitor cocktail. Store at 4 ^oC ( Detergent-free lysis buffer can be made up in advance and stored at 4 ^oC, without adding protease inhibitors. Protease inhibitors should be added on the day of use. Detergents must not be used in these buffers, as these would disrupt membrane integrity).
5. 6x concentrated lysis buffer (10 mL per experiment): 60 mM HEPES-HCl (pH 7.4), 12 mM EDTA, 12 mM magnesium acetate tetrahydrate, protease inhibitor cocktail. Store at 4 ^oC.
6. Ball bearing homogenizer (Isobiotec, Heidelberg, Germany).
7. Benzonase endonuclease, >90 % purity.
8. 1 and 2 mL syringes, without needles.

1. Mix iodixanol working solution and detergent-free lysis buffer to prepare 5 mL stocks with final concentrations of 8, 12, 16, 20, and 25 % iodixanol (w/v).
2. Take a 20 μL aliquot of each iodixanol solution and measure the refractive index with a handheld refractometer. If the value deviates from the expected by more than 0.25^oBx, then adjust with iodixanol working solution or lysis buffer and re-measure (The expected relationship between iodixanol concentration and refractive index can be established by measuring the refractive index for a series of iodixanol solutions and generating a calibration curve).
3. Using a 2 mL syringe and wide bore needle, add 2.2 mL of 8 % iodixanol solution to two 11 mL polyallomer Quick-Seal ultracentrifuge tubes.
4. Again using a wide bore needle, carefully layer 2.2 mL of 12 % iodixanol solution beneath the 8 % layer. Repeat this process for the 16 % and 20 % solutions, so that each tube contains approximately 8.8 mL in four visibly distinct iodixanol layers of increasing density towards the bottom of the tube (When layering solutions of iodixanol, care must be taken not to disrupt the gradient shape. Sequential underlaying of solutions is typically more robust than overlaying, as overlaying tends to result in solutions mixing. A wide bore needle and syringe allow easy access to the bottom of the gradient with minimal disruption, and decrease the probability of accidentally introducing air bubbles).
5. Place the tubes with layered iodixanol at 4 ^oC for at least 6 h, or at room temperature for at least 1 h, to allow the iodixanol layers to diffuse into a continuous gradient (The procedures outlined in this section of the protocol describe manually generating a pre-formed iodixanol gradient. If an automated gradient maker is available, then this can be used to generate the density gradient instead).

1. Centrifuge the cell lysate at 200 x g for 5 min to pellet unlysed cells and debris (If the nucleus is not an organelle of interest, then this centrifugation step can be performed at 2,000 x g instead of 200 x g . The post-nuclear supernatant may then be used for subcellular fractionation). Transfer the supernatant to a new tube and repeat this centrifugation step twice more.
2. Divide the supernatant equally into four 5.2 mL polyallomer ultracentrifuge tubes (Beckman). Using a wide bore needle, carefully underlay the supernatant in each tube with 0.8 mL 6 % iodixanol solution. Underlay once again with 0.8 mL 25 % iodixanol solution. The result should be three distinct layers in each tube, with the lysate on top. Balance pairs of ultracentrifuge tubes to within ∓10 mg with lysis buffer.
3. Centrifuge at 100,000 x g max for 90 min in a SW55Ti swinging bucket rotor. This centrifugation step pellets membranes out of the lysate and concentrates them at the interface of the two iodixanol layers.
4. Carefully pipette 0.5 mL from the middle of the supernatant layer in each tube, taking care not to disturb the gradient, and collect into a single 15 mL tube. Add 4 volumes of chilled acetone and place at -20 ^oC for several hours to allow proteins to precipitate. This fraction is strongly enriched with soluble cytosolic proteins.
5. Pipette membranes from the interface of the 6 and 25 % iodixanol layers and collect into a single 15 mL tube. If a membrane band has also collected at the interface of the supernatant and 6 % iodixanol layers, or pelleted to the bottom of the tube, then these may also be collected and pooled with the other membranes (Lower density membranes (e.g., plasma membrane fragments, endosomes etc.) may enrich at the interface of the supernatant and 6 % iodixanol layers, so collecting this secondary membrane layer will improve the final yields of these subcellular fractions).
6. Dilute any residual cytosolic material contaminating the membrane collection by addition of 5 volumes of lysis buffer, and mix with gentle pipetting or by inversion (Never mix membrane suspensions by vortexing, as this will disrupt membrane integrity). Transfer the membrane suspension into two or four 5.2 mL polyallomer ultracentrifuge tubes and balance to within ∓10 mg.
7. Centrifuge at 200,000 x g max for 60 min in a SW55Ti swinging bucket rotor. This centrifugation step pellets membranes to the bottom of the tube, and away from the residual cytosolic contamination in the initial membrane collection.
8. Decant the supernatant and resuspend the membrane pellets in a total volume of 2.0-2.5 mL 25 % iodixanol solution.
9. Using a wide bore needle and 2 mL syringe, carefully lay the membrane suspension underneath one of the two pre-formed iodixanol gradients ( see above ). Underlay the second pre-formed gradient with an approximately equivalent volume of 25 % iodixanol solution to act as a balancer during centrifugation. Balance the two tubes to within ∓10 mg.
10. Centrifuge at 100,000 x g max for 8 h in a VTi65.1 vertical rotor. During this ultracentrifugation step, membranes fl oat to and equilibrate at their respective buoyant densities.
11. Using an Auto Densi-Flow peristaltic pump with a meniscus tracking probe, collect 0.5 mL fractions of the density gradient into 1.5 mL polyallomer ultracentrifuge tubes (Beckman), from the top of the gradient downwards (A peristaltic pump with meniscus tracking probe allows for very reproducible gradient collection with minimal mixing. If such a device is unavailable, then fractions may be collected by pipetting from the top of the tube downwards, or by piercing the bottom of the tube and collecting fractions from the bottom of the gradient upwards).
12. Take a 20 μL aliquot of each fraction and measure the refractive index using a hand held refractometer to ascertain the final shape of the density gradient. The gradient should be almost linear, except for the densest 2 or 3 fractions, which may have higher iodixanol concentrations.
13. Add 0.8 mL lysis buffer to each gradient fraction, and mix gently by inversion. Balance pairs of tubes to within ∓10 mg. Samples can be kept at 4 ^oC overnight at this stage if necessary.
14. Centrifuge at 186,000 x g max for 20 min with a TLA-55 fixed angle rotor in a benchtop ultracentrifuge. During this centrifugation step membranes are pelleted from the iodixanol solution.
15. Examine each tube carefully to identify the membrane pellet. Pellets in the lower density fractions are typically smaller and more translucent, whereas those in the higher density fractions are larger and golden brown in color due to the abundance of cytochromes in the endoplasmic reticulum and mitochondria. Pipette 1.0 mL of the supernatant from the top of each tube and discard, taking care not to collect any membranes (For fractions from the bottom of the density gradient, membranes may not completely pellet due to the high iodixanol concentration. In such cases, remove as much of the supernatant as possible without disrupting the partially sedimented membranes). Add 1.0 mL lysis buffer to each tube to dilute the remaining iodixanol, and resuspend membrane pellets with gentle pipetting. Balance pairs of tubes to within ∓10 mg and repeat centrifugation step 14 .
16. Remove all of the supernatant from each tube and discard. Freeze membrane pellets to -80 ^oC for storage, or proceed with sample preparation (digestion, labeling).

Resources

1. Pierce Sample Preparation for Mass Spectrometry
2. Mass Spectrometry Sample Preparation Handbook (pdf)

References

1. Identification and quantification of N-linked glycoproteins using hydrazide chemistry, stable isotope labeling and mass spectrometry. Zhang H, Li XJ, Martin DB, Aebersold R.Nat Biotechnol. 2003 Jun;21(6):660-6. link
2. Dopaminergic modulation of the hippocampal neuropil proteome identified by bio-orthogonal non-canonical amino-acid tagging (BONCAT). J.J. Hodas et al., Proteomics 12, 2464-2476 (2012). link
3. Proteomics of human plasma: A critical comparison of analytical workflows in terms of effort, throughput and outcome. Loic Dayon, Martin Kussmann EuPA Open Proteomics Volume 1, 2013, Pages 8-16 12, 2464-2476 (2012). link
4. Shotgun Proteomics; Methods and Protocols Editors: Daniel Martins-de-Souza link