You need to ensure that you chelate Mg2+ (to inhibit kinases) with EDTA, add some EGTA to chelate Ca2+ (for calmodulin-dependent enzymes and proteases) and add phosphatase inhibitors to block dephosphorylation (e.g. 50 mM NaF, pervanadate, etc). The rest depends on the type of lysate you want to make (i.e. accessibility of the protein and sensitivity of the antibody) - RIPA is fairly stringent as it contains 1% NP-40, 1% cholate and 0.1% SDS. You can also lyse directly in 0.5% SDS (e.g. 5X Laemmli sample buffer), scape into a tube and boil for immediate denaturation. Otherwise, keep everything ice cold. As Alex said, avoid blocking with milk as that also screws up phosphotyrosine detection. Always control blot with antibodies against the same protein that are not phospho-specific to account for possible changes in expression. Also, remember that you cannot determine stoichiometry of phosphorylation (unless the blot is quantitative and you have a 100% phosphorylated control). Some phospho-specific antibodies have some background reactivity towards unphosphorylated epitopes.

I used Ripa buffer without any problem. Just need a good phosphatase inhibitor and protease inhibitor, and be the quickest possible between harvesting of live cells and protein samples ready (always working on ice, and snap freezing samples upon harvesting might help too). Also avoid using any milk (blocking buffer for WB for example) as it contains phosphatases.

Reference : http://www.abcam.com/ps/pdf/protocols/Western%20blotting%20of%20phospho-proteins%20protocol.pdf & http://webdoc.nyumc.org/nyumc/files/sun-lab/attachments/CPMB.Ch.18.Protein.Phosphorylation.pdf

You need to ensure that you chelate Mg2+ (to inhibit kinases) with EDTA, add some EGTA to chelate Ca2+ (for calmodulin-dependent enzymes and proteases) and add phosphatase inhibitors to block dephosphorylation (e.g. 50 mM NaF, pervanadate, etc). The rest depends on the type of lysate you want to make (i.e. accessibility of the protein and sensitivity of the antibody) - RIPA is fairly stringent as it contains 1% NP-40, 1% cholate and 0.1% SDS. You can also lyse directly in 0.5% SDS (e.g. 5X Laemmli sample buffer), scape into a tube and boil for immediate denaturation. Otherwise, keep everything ice cold. As Alex said, avoid blocking with milk as that also screws up phosphotyrosine detection. Always control blot with antibodies against the same protein that are not phospho-specific to account for possible changes in expression. Also, remember that you cannot determine stoichiometry of phosphorylation (unless the blot is quantitative and you have a 100% phosphorylated control). Some phospho-specific antibodies have some background reactivity towards unphosphorylated epitopes.

I think you have different options depending on what is your priority, -accessibility or sensitivity of Ab- as James already mentioned. I have been using RIPA w/o major problems. Be sure to chelate both Mg and Can and do not use milk in any step. Good luck!

Modified RIPA lysis buffer with phosphatase inhibitors (Beta-glycerophosphate, sodium ortho-vanadate and sodium pyrophosphate). 50mM Tris (pH-7.4), 150mM NaCl, 1mM EDTA, 1% NP-40, 0.25% sodium deoxycholate.

as long as phosphatase inhibitor is there, any buffer with 150mM sodium chloride in a physiological ph range will do

If all you want to look at is expression by western blot then this method works well. It is based on a very chaotropic lysis buffer called killer buffer; 2%SDS, 2M Urea, 14% sucrose, 1mM Sodium Fluoride,1mM Sodium Orthovanadate, 25mM Beta Glycero phosphate. It has the advantage that cells are lysed very quickly and proteins are dissociated and then denatured before phosphatases or proteinases get a chance to affect you protein of interest.

1. Remove the tissue culture medium, and rinse with ice cold PBS/sodium orthovanadate, incubating the last wash for 2 minutes to chill the samples. (1ml for a 24 well plate, 2ml for a 6 well plate, 12ml for a 90mm petri dish, 30ml for a 150mm petri dish).

2. Place a sheet of absorbant paper on the bench and remove the PBS.

3. Bang the plate or multiwall onto the paper to remove any excess PBS/ Sodiumorthovanadate without letting the paper touch the cells.

4. Add killer buffer to the cells (70μl for a 24 well plate, 100μl for a 6 well plate, 200μl for a 90mm petri dish, 400μl for a 150mm petri dish).

5. With a circular motion scrape the lysates around the plate (for the multiwell plates, use a shaft guard pipette with the end bent back, and a cell lifter for the plates) until the lysates is viscous- this is the genomic DNA suddenly released from the nucleus and stripped of all its histones..

6. Place the plate at an angle and gradually scrape the lysates to the lowest edge.

7. Using clean scissors cut the last 3-5mm off the end of the pipette tip at an angle and remove the lysates and transfer it to a Qiashredder column which has been placed in an Ependorff tube.

8. Centrifuge the Qiashredder columns for 2minutes at 6,000 rpm to shred the genomic DNA.

9. De-nature the protein by boiling the lysates at 100C°C for 3minutes.

10. The lysates can be stored on the bench for a while, or at -20°C for longer.

11. Measure the protein concentration using a BCA assay kit.

12. Add 2-Mercaptoethanol to 0.1% and Bromophenol blue to 0.0005% from stocks. The sucrose in the lysis buffer makes the protein extract sufficiently dense to sink in the wells of you gel.

Good luck

Lyse cells directly in 1x SDS-PAGE loading buffer.

I just directly diluted it by 4x SDS-PAGE loading buffer, and I always get what I want.

@Andrew Sunters, Your protocol looks good, and I have a few questions?

1) Do you have a reference about killer buffer? 2) Please look at this link, https://www.researchgate.net/post/How_to_stop_protein_aggregation, Stéphane Roche does not recommend to heat samples in low concentration of urea, which is different with step 9 of your protocol.

For SDS Lysis Buffer:

1% SDS, 50mM Tris (pH7.4), 150mM NaCl, 10mM EDTA, 1mM DTT, 10mM NaF, 1mM Na2VO3, 10mM beta glycerophosphate and 1X protease inhibitor.

For NP-40 lysis buffer you can use 1% NP-40 (instead of SDS as above). I have good experience with SDS lysis buffer. 

Do not wash the cells with PBS before lysis instead you can wash with fresh DMEM if cell debris you see on plate. Do not trypsinize the cells add lysis buffer directly to the plate keeping the plate on ice. Centrifuge, quick snap freeze the lysate and reduce any buffering time within experimental protocol from lysis to gel loading. And everything on ice.

And for WB, as Alex said No milk but 3% BSA for blocking and 1%BSA for antibody dilution. No PBST but TBST. 

Good Luck.