v468. chapter 4 WNT 细胞通路 检测细胞GSK3活性与表达水平
丁香园
9382
Measuring GSK3 Expression and Activity in Cells
Methods in Molecular Biology v468. chapter 4
AbstractGlycogen synthase kinase (GSK)-3 is a key signalling intermediate in the action of Wnts. This protein kinaseis ubiquitously expressed and has high inherent activity but is inhibited by activation of Wnt signalling oractivation of growth factor receptor tyrosine kinases (e.g. insulin, nerve growth factor [NGF], plateletderivedgrowth factor [PDGF], etc.). The degree of inhibition of GSK3 in cells treated with such reagents isdependent on the cell type and the stimulus used. Therefore, the ability to accurately measure GSK3 activityin cells is an important aspect of GSK3 research. The activity of GSK3 is reduced by posttranslational modification(phosphorylation) and this can be measured by immunoblot with specific reagents (indirect), or byimmunoprecipitation and assay (direct), as long as the modification is protected during these procedures.However, inhibition by phosphorylation is specific to cellular activation by growth factors and nutrients. Wnt inhibition of GSK3 does not involve phosphorylation of these residues on GSK3 and therefore it cannotbe measured using this modification. Currently, the simplest way to assess Wnt inhibition of GSK3 is tomonitor phosphorylation of specific GSK3 substrates in cells (e.g. b-catenin). Alternatively, Wnt inhibitionof GSK3 can be measured by partial purification of cellular GSK3 by ion exchange chromatography andassay of fractions or possibly by immunoprecipitation and assay. In this chapter, we demonstrate the use ofthe different approaches to measure GSK3 activity in SH-SY5Y cells, describe the best antibodies currentlyavailable, and discuss the potential drawbacks of each method.
Key words: GSK3 , Wnt , Assay , Phosphorylation , Growth factor .
1. Introduction
Glycogen synthase kinase (GSK)-3 is an evolutionarily conservedSer/Thr kinase that is ubiquitously expressed in all mammaliantissues and in all major subcellular organelles. Two isoforms areencoded by separate genes; GSK3b (chromosome 19q13.2)encodes a protein of 51 kDa, while GSK3b (chromosome 3q13.3)encodes a protein of 47 kDa (1 – 3) . The difference in size is dueto a glycine-rich N-terminal insertion in GSK3a. The bulk ofeach isoform is comprised of a central kinase domain (~350 amino acids) with 98% sequence homology between isoforms andis closely related to the cyclin-dependent kinase family. A splicevariant of GSK3b that contains a 13-amino acid insert in the catalyticdomain exhibits enhanced expression in the brain (4) .
GSK3 was first identified as the third kinase known to phosphorylateglycogen synthase (hence its name) as part of the insulinmediatedregulation of glucose metabolism (5) . Since then, it hasbeen shown to participate in many additional cellular functions,such as regulation of gene transcription, development, apoptosis,cell cycle, cytoskeletal dynamics, and molecular transport. GSK3activity is reported to be deregulated in diabetes and Alzheimer’sdisease (AD), and inhibitors are being assessed for treatment ofboth disorders (6) . To date, almost 50 substrates of GSK3 havebeen reported, with the majority being transcription factors, metabolicproteins, and cytoskeleton-associated proteins (7) .
GSK3 is one of the most unusual kinases of the 500 or soencoded in the human genome. Firstly, most (if not all) substratesrequire prior phosphorylation by another kinase beforethey can be efficiently phosphorylated by GSK3. This process isknown as “priming” and occurs 4 or 5 residues C-terminal tothe site phosphorylated by GSK3. Secondly, GSK3 is constitutivelyactive in cells under basal conditions. This is partly due toconstitutive phosphorylation of a conserved tyrosine residue onthe activation loop of the kinase domain (Tyr279 in GSK3a, andTyr216 in GSK3b), which is absolutely required for kinase activity.
Thirdly, phosphorylation of GSK3 at a conserved N-terminalserine residue inhibits its kinase activity. This phosphoserine actsas a pseudo-substrate and binds to the phosphate-binding pocketon GSK3, preventing interaction with primed substrates (8) .
Phosphorylation at this site is mediated by members of the AGCfamily of kinases (e.g. PKB, PKC, p70S6K, and p90RSK) andcommonly occurs downstream of growth factor and phosphatidylinositol3 kinase (PI3K) signalling. Activation of the canonicalWnt pathway also inhibits GSK3 activity, although this is notmediated by N-terminal phosphorylation but by protein–proteininteractions (9 – 11) (see earlier chapters).
Interestingly, most evidence suggests that growth factors andWnts regulate distinct pools of GSK3 in cells and thus each has adistinct set of downstream targets (Figs. 4.1 and 4.2 ). Part of this difference may be due to growth factor inhibition being mediatedthrough N-terminal phosphorylation of GSK3, resulting ininhibition of primed substrates, and Wnt inhibition focussing ona specific GSK3-containing complex. In addition, growth factorsregulate both GSK3a and GSK3b, while Wnts have so far onlybeen shown to regulate GSK3b (12) . Therefore the measurementof GSK3 activity in cells must take these differences into account.
There are several approaches available to measure GSK3activity in cells, however the approach chosen depends on thesignalling pathway being analysed. For example, inhibition byphosphorylation (following growth factor stimulation of cells)can be monitored by:
1. Immunoprecipitation of total cellular GSK3 and assay (seeSections 3.1 and 3.6 ). This will provide a “specific activity”
measurement relative to total protein in the cell, and is themost accurate measure of total GSK3 activity, or2. Immunoblot using antibodies to phospho-Ser21 (GSK3a)or phospho-Ser9 (GSK3b) (see Sections 3.1 – 3.4 ). This is acomparative analysis showing relative changes in the inhibitoryphosphorylation of GSK3 (13) . It should be noted thatit does not give an absolute measurement of GSK3 activity,since the stoichiometry of phosphorylation cannot be assessedby this method. Immunoblot using antibodies to phospho-Tyr279/216 can assess alterations in activity independent ofthe N-terminal phosphorylation, although phosphorylation atthis site is most likely to be constitutive (14) .
3. Immunoblot using antibodies to known substrates of GSK3,for example collapsin response mediator protein (CRMP) -2(Fig. 4.2 and ref. (15) ).
On the other hand, assessment of Wnt regulation of GSK3 ismore complex. Four methods have been used previously:
1. Immunoprecipitation of total cellular GSK3 and assay asabove, although this may be a very transient effect as we(Fig. 4.3 ) and others (10 , 12) do not see any change in GSK3specific activity beyond 30 min exposure of cells to Wnt.
Fig. 4.1. Wnt3a regulates GSK3-mediated phosphorylation of b-catenin, but not CRMP2. A Lysates of SH-SY5Y cellsincubated with conditioned medium from control (–) or Wnt3a-expressing (+) L cells (1 h) were subjected to SDS-PAGEand transferred to nitrocellulose membranes. Membranes were probed with antibodies that recognise GSK3a/b whenphosphorylated at Ser21/9, Tyr279/216, or total GSK3a/b. The band intensities were quantitated and the ratios of phosphorylatedGSK3a/b to total GSK3a/b are shown (control lysates normalised to 100%). Error bars indicate the range ofduplicate experiments. B Same as ( A ), except membranes were probed with antibodies that recognise phosphorylatedb-catenin (Thr41/Ser37/Ser33), total b-catenin, phosphorylated CRMP2 (Thr514/509), or total CRMP2. n.s. , non-significant~undefined p <0.05 (Student’s t test) .
Fig. 4.2. IGF1 regulates GSK3-mediated phosphorylation of CRMP2, but not b-catenin. A Lysates of SH-SY5Y cells incubatedwithout (–) or with (+) IGF1 (50 ng/mL, 1 h) were subjected to SDS-PAGE and transferred to nitrocellulose membranes. Membranes were probed with antibodies that recognise GSK3a/b when phosphorylated at Ser21/9, Tyr279/216,or total GSK3a/b. The band intensities were quantitated and the ratios of phosphorylated GSK3a/b to total GSK3a/b areshown as a graph (control lysates normalised to 100%). Error bars indicate the range. B Same as ( A ), except membraneswere probed with antibodies that recognise phosphorylated b-catenin (Thr41/Ser37/Ser33), total b-catenin, phosphorylatedCRMP2 (Thr514/509), or total CRMP2. n.s. , non-significant; * p < 0.05 (Student’s t test) .
Fig. 4.3. IGF1, but not Wnt3a, reduces cellular GSK3b activity. SH-SY5Y cells werestimulated with IGF1 or Wnt3a (1 h), then harvested in cell lysis buffer. GSK3b wasimmunoprecipitated from lysates and its activity was measured in an in vitro kinaseassay using pGS2 peptide. GSK3b activity is presented in a graph as a percentage ofuntreated cells (normalised; 100% activity from control cells equates to 52 pmol/min/mg). n.s. , non-significant; *, p < 0.05 (Student’s t test) .
2. Comparative analysis of phosphorylation of the Wnt targetprotein b-catenin by immunoblot (see Sections 3.1 – 3.5 ); or
3. Partial purification of cellular GSK3 by cation exchange chromatography(12) , followed by assay (see Section 3.6 ). Thisapproach assumes that the inhibitory protein complex is maintainedduring the purification procedure and that a substantialproportion of cellular GSK3 is inhibited (hence, this is subjectto the cell system used and variations in protein purificationthat may disrupt the adenomatous polyposis coli [APC]
complex). As with the immunoprecipitation assay above, thismeasurement may be very transient; or
4. Immunoprecipitation of the Axin–APC complex rather thantotal GSK3 (16 , 17) followed by assay. However, this approachassumes that this is the only GSK3 complex regulated by Wntand that any inhibitory interactions are maintained duringthe isolation procedure. Interestingly, this method predictedrelatively poor inhibition of GSK3 by Wnt compared with theother approaches.
In this chapter, we provide technical details of the immunoprecipitationand assay of GSK3 as well as immunoblot analysis ofGSK3 and its substrates.
2. Materials
2.1. Cell Culture and Lysis
1. SH-SY5Y cells (ATCC #CRL-2266) were maintained in Dulbecco’sModified Eagle’s Medium (DMEM; Cambrex , East Rutherford,NJ) supplemented with 10% (v/v) foetal bovine serum (FBS; Gibco; Invitrogen, Carlsbad, CA), penicillin/streptomycin(Gibco) and 10 μM retinoic acid to induce neuronaldifferentiation (see Note 1 ).
2. Insulin-like growth factor (IGF)-1 (Invitrogen) was dissolvedin phosphate-buffered saline (PBS) containing 0.1% (w/v)bovine serum albumin (BSA) at a concentration of 50 μg/mLand stored in aliquots at –20°C (see Note 2 ).
3. Mouse L cells that express and secrete Wnt3a (ATCC, Manassas,VA; #CRL-2647) and the control L cells that do notsecrete Wnt3a (ATCC #CRL-2648) are grown in DMEMcontaining 10% (v/v) FBS and penicillin/streptomycin. Mediais collected from these cells and stored at 4°C for up to 1month or at -20°C for longer periods (see Note 3 ).
4. Cell lysis buffer: 50 mM Tris-HCl, pH 7.5, 1% (v/v) TritonX-100, 0.27 M sucrose, 1 mM EDTA, 0.1 mM EGTA, 1 mMsodium orthovanadate, 50 mM sodium fluoride, 5 mM sodiumpyrophosphate, 0.1% (v/v) b-mercaptoethanol, and completeprotease inhibitor cocktail (Roche, Basel, Switzerland).
5. Bradford reagent for determining lysate protein concentrationsis from Sigma-Aldrich (St. Louis, MO; #B6916).
2.2. Sodium Dodecyl Sulphate (SDS)- Polyacrylamide Gel Electrophoresis (PAGE) and Electrotransfer
1. Pre-cast NuPAGE Bis-Tris 4–12% polyacrylamide gels fromInvitrogen (see www.invitrogen.com) provide high-qualityand reproducible separation of proteins, but are expensive,are not suitable for integral membrane proteins, and requirespecialised tanks and buffers for best performance. StandardSDS-PAGE is an alternative to this system (18) .
2. Stock solutions of morpholinepropane sulphonic acid (MOPS)electrophoresis running buffer (20×), transfer buffer (20×),antioxidant solution (1,000×), SDS sample buffer (4×), andSeeBlue Plus2 molecular weight markers are available fromInvitrogen and are optimised for the NuPAGE system. Thesereagents are based on standard SDS-PAGE reagents ( RunningBuffer , 0.1 M glycine, 25 mM Tris-HCl, pH 7.4, and0.05% (w/v) SDS; Transfer Buffer , 25 mM Tris-HCl, pH 7.4,192 mM glycine, and 20% (v/v) methanol; Sample Buffer , 25mM Tris-HCl, pH 6.8, 2% (w/v) SDS, 10% (v/v) glycerol,and 0.1 μg/mL Bromophenol Blue).
3. Nitrocellulose membranes are available from Amersham (HybondC-extra, #RPN303E; GE Healthcare, Piscataway, NJ).
2.3. Immunoblotting
1. Many different types of phosphorylation-dependent andphosphorylation-independent antibodies for GSK3 are commerciallyavailable and are listed in Table 4.1 . For the experimentsshown in this chapter, the following antibodies wereused; GSK3α/b (pS21/9) (#9331; Cell Signaling Technology, Danvers, MA), GSK3a/b (pY279/216) (#612312; BDTransduction Laboratories, San Jose, CA) and GSK3a/b (#05-903; Upstate; Millipore, Billerica, MA) (see Note 4 ).
3. Blocking buffer consists of 5% (w/v) dried skim milk in Trisbufferedsaline/Tween-20 (TBS-T) buffer (50 mM Tris-HCl,pH 7.5, 150 mM NaCl, and 0.05% (v/v) Tween-20).
4. All antibodies are diluted 1:1,000 in 1% (w/v) dried skim milkin TBS-T, except the CRMP2 antibody (1:250) (see Note 5 ).
5. Rabbit anti-mouse (#31450), sheep anti-rabbit (#31460)and donkey anti-sheep (#31480) horseradish peroxidaseconjugatedsecondary antibodies are available from Pierce(Rockford, IL).
6. Enhanced chemiluminescence (ECL) reagent (#RPN2106)and X-ray films (#RPN3103K) are available from Amersham.
2.4. Image Analysi and Quantitation
1. Images are digitised by scanning and densitometry using Aidaquantitation software. Data analysis is performed using MicrosoftExcel software.
2. Images are prepared using Adobe Photoshop and MicrosoftPowerpoint software.
2.5. Immunoprecipitation Kinase Activity Assays
1. The anti-GSK3b monoclonal antibody is from BD TransductionLaboratories (#610201) and protein G-Sepharose is fromAmersham (#17-0618-01).
2. Radioactive [g- 32 P]ATP is from Amersham (#PB10218), P81paper from Whatman (#3698-915; Maidstone, UK) and orthophosphoricacid from BDH AnalaR (#10173CD; VWR, Lutterworth,UK). Scintillation fluid is from Perkin Elmer (#6013371;Waltham, MA) and a Beckman LS-6500 scintillation counter isused to detect radioactivity (Beckman Coulter, Fullerton, CA).
3. Kinase buffer consists of 50 mM Tris-HCl, pH 7.5, 150 mMNaCl, 0.03% (v/v) Brij-35, and 0.1% (v/v) b -mercaptoethanol.
4. The 10× [g- 32 P]-MgATP stock solution contains 100 mMMgCl 2 , 1 mM un-labelled ATP, and approximately 30 MBq/mL [g- 32 P]ATP.
5. Phospho-glycogen synthase 2 (pGS2) peptide is available fromUpstate (#12-241) and the pGS2 10× stock solution contains200 μM pGS2 in kinase buffer.
3. Methods
3.1. Cell Stimulation and Lysis
1. SH-SY5Y cells are grown on 10-cm tissue culture plates inmedium containing retinoic acid for 4 days to induce neuronaldifferentiation. On the fourth night, remove the medium,rinse the cells once in pre-warmed serum-free DMEMcontaining penicillin/streptomycin and retinoic acid, andincubate in this medium overnight to serum-starve the cells.
This maximally induces GSK3 activity.
2. The following morning, stimulate the cells by adding 50 ng/mL IGF1 or carrier solution for the desired time(s) (here, 1 hat 37°C). Alternatively, add an equal volume of conditionedmedium from control or Wnt3a-expressing L cells to themedium already in the plate for the desired time(s) (here, 1 hat 37°C).
3. Harvest the cells by placing the plates on ice, aspirate themedium, rinse the cells once in ice-cold PBS, and scrape thecells directly into 0.2 mL of ice-cold cell lysis buffer. Leave onice for 5 min and then centrifuge the samples (14,000× g , 10min) to pellet insoluble material. Immediately snap-freeze thesupernatant (lysate) in liquid nitrogen and discard the pellet.
3.2. Preparation for SDS-PAGE
1. Determine the protein concentrations of each sample usingthe Bradford assay (19) . Briefly, add various amounts of lysate(usually 0.1–10 μL in a total volume of 100 μL of water) to1 mL of Coomassie (Bradford) reagent, vortex, and incubateat room temperature for 10 min. Vortex the solutions again,then measure the absorbance using a spectrophotometer atl max 595 nm. Compare the absorbance of each sample solutionto a standard curve of BSA (serial dilution: 0 to 10 μg)in order to establish the concentration of protein in eachsample. All samples and standards should be assayed in triplicate.
2. As a general rule, it is best to load equal volumes of samplesas well as amounts of total protein for each sample to theirrespective lanes on the gel. For example, equal amounts oftotal protein from each sample in Fig. 4.1 (160 μg, withvolumes ranging from 136 to 231 μL) are diluted with differentamounts of cell lysis buffer to give the same finalvolume (240 μL). Then, 80 μL of 4× SDS sample loadingbuffer containing 40 mM dithiothreitol (DTT) is added toeach sample, mixed, and heated to 70°C for 15 min. Everydenatured sample therefore contains 0.5 mg/mL total protein,and every lane is loaded with the same protein amountand volume .
3.3. SDS-PAGE and Protein Transfer
1. Rinse the wells of the pre-cast polyacrylamide gels with waterto remove unpolymerised polyacrylamide, then insert the gelsinto the electrophoresis tank and lock them in place.
2. Add running buffer (1×) to the inner and outer chambers ofthe tank.
3. Load molecular weight markers (e.g. SeeBlue Plus 2–5 μL)into lane 1 of the gel using gel-loading pipette tips, then loadequal volumes of each sample into the other lanes (usually4–40 μL containing 2–20 μg of total protein).
4. Connect the tank to a power supply and run the gels at 120 Vfor 20 min at room temperature, then at 180 V until the gelfront reaches the exposed strip of polyacrylamide at the bottomof the gel.
5. While the gel is running, prepare the equipment and reagentsfor the transfer step. Immerse the sponges in transferbuffer and remove any air bubbles by tapping with a spatula.
Cut squares of blotting paper and nitrocellulose membranesslightly smaller in size than the sponges, label the membraneswith a small mark in a corner using a pencil, and immersethem all in transfer buffer.
6. Remove the gel from the electrophoresis tank and open theplastic casing of the gel using a spatula. Carefully remove thegel and submerge it in transfer buffer.
7. Assemble the transfer cassette in the following order: sponge,blotting paper, gel, membrane, blotting paper, and sponge.
Place the cassette into the electrophoresis tank with the geltowards the negative terminal and the membrane towards thepositive terminal. When a current is applied, the negativelycharged SDS–protein complexes will move to the membrane.
Add transfer buffer, connect to a power source, and transferthe proteins from the gel to the membrane at 35 V for 1–2 h.
8. When the transfer is complete, remove the membrane fromthe cassette and incubate it in blocking buffer for 1 h at roomtemperature or overnight at 4°C (enough volume to ensurethat no part of the membrane becomes dry).
3.4. Immunoblotting for GSK3 and GSK3 Substrates
1. Discard the blocking buffer and wash the membranes 3× 10min in TBS-T.
2. Incubate the membranes with the primary antibody diluted in1% (w/v) milk in TBS-T (see Note 5 ) with constant shakingor rotating to ensure even coverage of the membrane with theantibody. This step is usually performed at 4°C overnight, butcan also be performed for shorter periods at room temperature(at least 1 h).
3. Remove the membranes from the primary antibody solutionand wash them 3× 10 min in TBS-T. Most primary antibody solutions can be re-used three to five times without loss ofsensitivity or selectivity, but should be stored at 4°C for upto 1 month (add 0.02% [w/v] sodium azide to prevent bacterialgrowth). Dilute the appropriate secondary antibody ina solution containing 1% (w/v) milk in TBS-T according tothe manufacturer’s instructions (usually 1:2,500–1:10,000).
Incubate the membranes in this solution for 1 h at room temperaturewith shaking or rotation.
4. Discard the secondary antibody solution and wash the membranes3× 10 min in TBS-T.
5. The ECL reagents A and B should be removed from the fridgeto allow them to equilibrate to room temperature before use.
In a dark room under a safety light, mix equal volumes ofA and B (e.g. 1 mL + 1 mL for a single membrane) for nomore than 1 min. Drain excess TBS-T from the membranesand place them on a flat, clean surface. Pipette enough of themixed ECL reagent to cover the entire membrane and let itstand for 1 min.
6. Quickly drain the ECL reagent from the membrane and placeit in inside Saran wrap in an X-ray film cassette.
7. Place an X-ray film onto the membrane (initially try 1 min)then develop the film (see Note 6 ). While developing the firstfilm, a second and third exposure to film can be performed fora longer or shorter period of time to provide different exposuresof the image. This is important in order to establish thelinearity of the ECL reaction for each sample with each antibody.
8. Membranes can be re-probed with different antibodies (seeNote 7 ).
3.5. Image Analysis and Quantitation of Immunoblots
1. Choose an exposure with an intensity in the linear range of theECL/immunoblotting technique. As a general rule, bandsshould be solid and black (light and grey suggests bands areunder-exposed) and should contain sharp edges (blurred/fuzzy edges suggests bands are overexposed). Over-sized bands(“ballooned”) also suggests that bands are overexposed.
2. Digitise the X-ray films containing images of the immunoblotsby scanning. Suitable software (such as Aida) can be used toquantify band intensity (see Note 8 ).
3. To determine changes in phosphorylation of a protein, calculatethe ratio of phosphorylated protein to total protein foreach sample (e.g. divide the densitometric value of pSer21/9-GSK3 by the value for the corresponding total GSK3a/b). The membranes for phospho and total GSK3 analyses shouldbe generated using the same denatured lysate solution andperformed at the same time.
4. Figure 4.1a shows analysis of phosphorylation of GSK3 isoformsat Ser21/9 and Tyr279/216 in SH-SY5Y cells followingstimulation with Wnt3a. Representative blots andquantified ratios of phosphorylation relative to total proteinare provided. Figure 4.1b shows analysis of the same celllysates using phospho-β-catenin and p-CRMP2, two establishedtargets of GSK3. Note that only the phosphorylationof β-catenin is reduced following Wnt stimulation, thereforephosphorylation of GSK3 or CRMP2 cannot be used to monitorWnt regulation of GSK3 activity in cells. In contrast, IGF1stimulation of these cells (Fig. 4.2a ) induces phosphorylationof Ser21/9 but does not affect Tyr279/216 phosphorylation,and reduces phosphorylation of CRMP2 but does not affectphosphorylation of β-catenin (Fig. 4.2b ). Therefore, IGF1regulation of GSK3 can be visualised by changes in GSK3and/or CRMP2 phosphorylation.
3.6. Immunoprecipitation Kinase Activity Assays (See Note 9)
1. Pre-incubate anti-GSK3β (5 μg) monoclonal antibody withprotein G-Sepharose (400 μL of a 50% slurry) in kinasebuffer at 4°C for 1 h to conjugate the antibody to the proteinG-Sepharose beads.
2. Centrifuge the beads at 6,000× g for 1 min, discard the supernatant,and wash the beads two times in ice-cold PBS toremove unbound antibody. Resuspend the beads in 250 μL ofice-cold PBS.
3. Incubate 100 μL of beads with 400 μg of lysate in a totalvolume of 1 mL of cell lysis buffer (4°C, 1 h) with shaking orrotating.
4. Centrifuge at 6,000× g for 1 min, wash once in cell lysis buffercontaining 0.5 M NaCl, once in cell lysis buffer, then resuspendedin 190 μL of kinase buffer (all performed on ice).
5. Separate the beads into four aliquots of 40 μL and add 5 μLof pGS2 peptide solution while on ice.
NOTE: All steps from this point involve the use of radioactive[ γ - 32 P]-MgATP, so appropriate safety precautions and disposalof radioactive waste should be strictly employed.
6. To start the kinase reaction, add 5 μL of 10× [γ- 32 P]-MgATPsolution and incubate on an immunoprecipitation shaker at30°C for 15 min.
7. To terminate the reaction, spot 40 μL onto 1.5×1.5-cm squaresof P81 paper (marked numerically in pencil) and immediatelyimmerse in 75 mM orthophosphoric acid.
8. Wash the P81 papers 3× 10 min in 75 mM orthophosphoricacid. Use a magnetic stirrer to continually agitate the acid toensure efficient washing of the papers.
9. Discard the final wash and briefly immerse the papers in acetone.
Discard the acetone and air-dry the papers (this can beexpedited using a hair dryer).
10. Place each paper in individually labelled tubes. Add 2 mLof scintillant to each tube and measure the radioactivity.
Alternatively, the radioactivity can be measured by Cerenkovcounting (if an appropriate detector is available), which doesnot require the addition of scintillation fluid.
11. Calculation of GSK3 specific activity requires the measurementof the specific activity of the ATP. 1 μL of 10× [γ- 32 P]-MgATPcontains 1 nmole of ATP, therefore the counts per minute(cpm) in 1 μL of this solution (counted in 2 mL of scintillantas above) provides the counts per minute per nanomole forthe substrate (usually around 200,000 cpm/nmole ). Thismust be counted simultaneously with the papers. Therefore,if GSK3 precipitated from 100 μg of cell protein produces10,000 cpm in 15 min, then the specific activity of the GSK3in that cell lysate is 10,000/200,000 cpm/nmole phosphatetransferred per 15 min/0.1 mg cell protein under the assayconditions used; or 0.0333 nmoles/min/mg or 33 pmoles/min/mg.
Use of this assay allows direct comparison of GSK3 activitiesbetween experiments and cell systems from different days and indifferent locations.
12. Figure 4.3 demonstrates that 1 h exposure to IGF1 producesa 33% reduction in the specific activity of total GSK3in SH-SY5Y cells, while there is no measurable effect of 1 hexposure of Wnt3a on total GSK3 activity (it should be notedthat others have reported a 40–50% reduction upon 10 minexposure to Wnt proteins in other systems (10 , 12) ).
3.7. Comparison of Methodology for Analysis of GSK3 Activity
As described above, care must be taken in selection of the mostappropriate method when analysing GSK3 activity. Upon IGF1stimulation, GSK3 Ser21/9 phosphorylation increases threefold(Fig. 4.2a ), and this equates to a drop in phosphorylation of theGSK3 substrate CRMP2 of 44% (Fig. 4.2b ) and a reductionin GSK3 specific activity of 33% (Fig. 4.3 ). If all other regulationof GSK3 is constant (e.g. Tyr279/216 phosphorylation, Fig. 4.2a )then the threefold induction of Ser21/9 phosphorylation predictsan increase of stoichiometry at these sites from 0.1 to 0.3 mol/mol,since 67% of GSK3 remains active. Interestingly, the analysis ofCRMP2 phosphorylation is very close to that of the specific activitymeasurement. This would suggest that the phosphorylation ofThr514/509 of CRMP2 by GSK3 is very sensitive to changes incellular GSK3 activity. Each proposed substrate of GSK3 has to beanalysed in combination with GSK3 activity in order to establishwhether it is representative of changes in GSK3 activity.
In contrast, Wnt3a treatment of cells for 1 h does not alter thespecific activity of GSK3 (Fig. 4.3 ). This is similar to experimentswhere GSK3β associated to axin was isolated and measured followingchallenge of cells with Xwnt8 (17) . However, partial purificationof GSK3 on a MonoS column followed by in vitro assaysuggested that Drosophila Wingless (Wg) could reduce total cellularGSK3 activity in fibroblasts by as much as 50% (12) . It islikely that the discrepancy is due to a transient inhibition of GSK3by Wnt (10 , 12) . However, the phosphorylation of the knownWnt/GSK3 pathway target β-catenin is almost completely lostand this reduction is sustained even after 1 h exposure to Wnt(Fig. 4.1b ). This strongly suggests that either only a small poolof Wnt-responsive GSK3 is present in SH-SY5Y cells or that theinhibition of GSK3 is transient but the dephosphorylation ofβ-catenin is sustained. Either way, β-catenin phosphorylation ispurely a qualitative method for measuring regulation of GSK3 byWnts in cells, and does not reflect either the degree or durationof regulation of GSK3 activity.
4. Notes
1. GSK3 is ubiquitously expressed. Therefore, the protocols describedhere could be adapted for any cell line/tissue.
2. GSK3 is inhibited by all growth factor receptor tyrosine kinasesinvestigated (e.g. insulin, epidermal growth factor [EGF], NGF,PDGF, hepatocyte growth factor [HGF], brain-derived neurotrophicfactor [BDNF], neurotrophin [NT]-3). Here, wehave focussed our study on IGF1.
3. It is generally considered that any Wnt molecules that regulateβ-catenin activity inhibit GSK3. Reduced cellular GSK3activity has been demonstrated for Drosophila Wingless (12) ,Xenopus Wnt8 (17) , and mammalian Wnt1 (10) . We haveused mammalian Wnt3a in our studies.
4. Many commercial antibodies that recognise phosphorylatedand non-phosphorylated epitopes on GSK3 are listed in Table4.1 . We have not systematically tested all of the antibodieslisted. However, of the antibodies that we have used, we recommendthe antibodies listed in Section 2.3.1 .
5. Instead of milk in the blocking and antibody solutions, someantibodies work best in 1% BSA, especially antibodies thatdetect phospho-tyrosine epitopes. Also, some companies recommenda higher concentration of milk than 1%. For example,Cell Signaling Technology recommends the use of thepSer21/9-GSK3 antibody in 5% milk. However, our experience shows that its sensitivity is improved in 1% milk without lossof specificity.
6. The luminescence emitted by the ECL/horseradish peroxidase(HRP)-conjugated secondary antibodies decreases withtime, so it is best to expose the membranes to film immediatelyafter addition of the ECL reagent, especially when usingantibodies with low sensitivity.
7. Stripping then re-probing the nitrocellulose membranes withanother antibody is an option when samples are in short supply,however in general it is not recommended. The strippingprocedure is often inconsistent across the surface of the membrane,resulting in unreliable band intensities across the lanes, aswell as higher-than-normal background. In addition, strippedmembranes cannot be re-probed with phospho-specific antibodies,since the stripping solution also removes phosphatemoieties from proteins on the membrane. As an alternative,membranes should be washed to remove the ECL reagent,re-blocked, and incubated with an antibody that recognisesan antigen of a molecular weight that is sufficiently differentto that of the first antibody/antigen. The resultant film willcontain bands for both the first and second antibodies used onthis membrane, but the large difference in molecular weightwill easily distinguish the two antigens.
8. The immunoblotting approach described here is semi-quantitative.
Provided the intensities of each band are in the linear range ofdetection, this will provide comparative data to bands withinthe same analysis (membrane). If absolute quantification isrequired, a standard curve of known amounts of purified antigenmust be analysed, preferably on the same gel but at leastdeveloped at the same time.
9. The immunoprecipitation assay of GSK3 is described. Howeverthere are variations on this procedure that have proved useful. For example Cook et al. partially purified GSK3 on a MonoSFPLC column prior to in vitro assay of the fractions against thepGS2 peptide (12) . This could be used to measure either thegrowth factor or Wnt regulation of GSK3. Secondly, Cross et al. developed a procedure to measure GSK3 in a crude cell lysate byperforming the in vitro GSK3 assay (see Section 3.6 ) with thedephosphorylated (unprimed) GS (glycogen synthase) peptidesubstrate as well as the pGS2 (phospho-glycogen synthase peptide2). peptide (20) . The difference in the two measurementsshould be GSK3 activity. However, it should be noted that thepeptide has several phosphorylatable residues and therefore itmay be targeted by other kinases in the lysate (reducing the signal-to-noise ratio or interfering with GSK3 phosphorylation). In addition, the large number of kinases likely to be active (andpotentially affected by growth factor stimulation) can deplete the ATP and reduce the linearity of the assay. Finally, the otherkinases in the assay may indirectly regulate GSK3 activity duringthe in vitro assay, hence this assay may not truly represent theactivity state of GSK3 in cells. A third modification of the immunoprecipitationassay involves pre-incubating the GSK3 with orwithout the phosphatase PP2A (21) . This treatment removesthe inhibitory phosphorylation at Ser21/9 thereby fully activatingthe enzyme. Comparison of GSK3 activity with or withoutpre-treatment with PP2A provides an estimate of the percentageof active versus inactive (Ser21/9-phosphorylated) GSK3in the cells under each experimental condition.
Acknowledgments
This work was supported by the Alzheimer’s Research Trust andDiabetes UK.
References
1. Woodgett, J. R. (1990) Molecular cloning and expression of glycogen synthase kinase- 3/factor A. EMBO J. 9, 2431–2438.
2. Hansen, L., Arden, K. C., Rasmussen, S. B.,Viars, C. S., Vestergaard, H., Hansen, T., et al. (1997) Chromosomal mapping and mutational analysis of the coding region of the glycogen synthase kinase-3alpha and beta isoforms in patients with NIDDM. Diabetologia 40, 940–946.
3. Shaw, P. C., Davies, A. F., Lau, K. F., Garcia-Barcelo, M., Waye, M. M., Lovestone, S., et al. (1998) Isolation and chromosomal mapping of human glycogen synthase kinase-3 alpha and -3 beta encoding genes.Genome 41, 720–727.
4. Mukai, F., Ishiguro, K., Sano, Y., and Fujita, S. C. (2002) Alternative splicing isoform of tau protein kinase I/glycogen synthase kinase 3beta. J. Neurochem. 81, 1073–1083.
5. Embi, N., Rylatt, D. B., and Cohen, P. (1980) Glycogen synthase kinase-3 from rabbit skeletal muscle. Separation from cyclic-AMP-dependent protein kinase and phosphorylase kinase. Eur. J. Biochem. 107, 519–527.
6. Cohen, P. and Goedert, M. (2004) GSK3 inhibitors: development and therapeutic potential. Nat. Rev. Drug Discov. 3, 479–487.
7. Frame, S. and Cohen, P. (2001) GSK3 takes centre stage more than 20 years after its discovery. Biochem. J. 359, 1–16.
8. Frame, S., Cohen, P., and Biondi, R. M. (2001) A common phosphate binding site explains the unique substrate specificity of GSK3 and its inactivation by phosphorylation. Mol. Cell 7, 1321–1327.
9. Thomas, G. M., Frame, S., Goedert, M., Nathke, I., Polakis, P., and Cohen, P. (1999) A GSK3-binding peptide from FRAT1 selectively inhibits the GSK3-catalysedphosphorylation of axin and beta-catenin. FEBS Lett. 458, 247–251.
10. Ding, V. W., Chen, R. H., and McCormick, F. (2000) Differential regulation of glycogensynthase kinase 3beta by insulin and Wnt signaling. J. Biol. Chem. 275, 32475– 32481.
11. Liu, X., Rubin, J. S., and Kimmel, A. R. (2005) Rapid, Wnt-induced changes in GSK3beta associations that regulate betacateninstabilization are mediated by Galpha proteins. Curr. Biol. 15, 1989–1997.
12. Cook, D., Fry, M. J., Hughes, K., Sumathipala,R., Woodgett, J. R., and Dale, T. C. (1996) Wingless inactivates glycogen synthasekinase-3 via an intracellular signalling pathway which involves a protein kinase C. EMBO J. 15, 4526–4536.
13. Sutherland, C., Leighton, I. A., and Cohen,P. (1993) Inactivation of glycogen synthase kinase-3 beta by phosphorylation: new kinase connections in insulin and growthfactorsignalling. Biochem. J. 296 (Pt 1), 15–19.
14. Cole, A., Frame, S., and Cohen, P. (2004)Further evidence that the tyrosine phosphorylationof glycogen synthase kinase-3 (GSK3)in mammalian cells is an autophosphorylationevent. Biochem. J. 377, 249–255.
15. Cole, A. R., Knebel, A., Morrice, N. A., Robertson, L. A., Irving, A. J., Connolly, C. N. et al. (2004) GSK-3 phosphorylation of the Alzheimer epitope within collapsin response mediator proteins regulates axon elongation in primary neurons. J. Biol. Chem. 279, 50176–50180.
16. Hart, M. J., de los, S. R., Albert, I. N.,Rubinfeld, B., and Polakis, P. (1998)Downregulation of beta-catenin by humanAxin and its association with the APC tumor suppressor, beta-catenin and GSK3 beta. Curr. Biol. 8, 573–581.
17. Itoh, K., Krupnik, V. E., and Sokol, S. Y. (1998) Axis determination in Xenopus involves biochemical interactions of axin, glycogen synthase kinase 3 and beta-catenin. Curr. Biol. 8, 591–594.
18. Laemmli, U. K. (1970) Cleavage of structuralproteins during the assembly of thehead of bacteriophage T4. Nature 227,680–685.
19. Bradford, M. M. (1976) A rapid and sensitivemethod for the quantitation of microgramquantities of protein utilizing theprinciple of protein-dye binding. Anal. Biochem. 72, 248–254.
20. Cross, D. A., Culbert, A. A., Chalmers, K. A., Facci, L., Skaper, S. D., and Reith, A. D. (2001) Selective small-molecule inhibitors of glycogen synthase kinase-3 activity protect primary neurones from death. J. Neurochem. 77, 94–102.
21. Cross, D. A., Alessi, D. R., Vandenheede,J. R., McDowell, H. E., Hundal, H. S., and Cohen, P. (1994) The inhibition of glycogen synthase kinase-3 by insulin or insulin-like growth factor 1 in the rat skeletal muscle cell line L6 is blocked by wortmannin, but not by rapamycin: evidence that wortmannin blocks activation of the mitogen-activated protein kinase pathway in L6 cells between Ras and Raf. Biochem. J. 303 (Pt 1), 21–26.
References
1. Woodgett, J. R. (1990) Molecular cloning and expression of glycogen synthase kinase- 3/factor A. EMBO J. 9, 2431–2438.
2. Hansen, L., Arden, K. C., Rasmussen, S. B.,Viars, C. S., Vestergaard, H., Hansen, T., et al. (1997) Chromosomal mapping and mutational analysis of the coding region of the glycogen synthase kinase-3alpha and beta isoforms in patients with NIDDM. Diabetologia 40, 940–946.
3. Shaw, P. C., Davies, A. F., Lau, K. F., Garcia-Barcelo, M., Waye, M. M., Lovestone, S., et al. (1998) Isolation and chromosomal mapping of human glycogen synthase kinase-3 alpha and -3 beta encoding genes.Genome 41, 720–727.
4. Mukai, F., Ishiguro, K., Sano, Y., and Fujita, S. C. (2002) Alternative splicing isoform of tau protein kinase I/glycogen synthase kinase 3beta. J. Neurochem. 81, 1073–1083.
5. Embi, N., Rylatt, D. B., and Cohen, P. (1980) Glycogen synthase kinase-3 from rabbit skeletal muscle. Separation from cyclic-AMP-dependent protein kinase and phosphorylase kinase. Eur. J. Biochem. 107, 519–527.
6. Cohen, P. and Goedert, M. (2004) GSK3 inhibitors: development and therapeutic potential. Nat. Rev. Drug Discov. 3, 479–487.
7. Frame, S. and Cohen, P. (2001) GSK3 takes centre stage more than 20 years after its discovery. Biochem. J. 359, 1–16.
8. Frame, S., Cohen, P., and Biondi, R. M. (2001) A common phosphate binding site explains the unique substrate specificity of GSK3 and its inactivation by phosphorylation. Mol. Cell 7, 1321–1327.
9. Thomas, G. M., Frame, S., Goedert, M., Nathke, I., Polakis, P., and Cohen, P. (1999) A GSK3-binding peptide from FRAT1 selectively inhibits the GSK3-catalysedphosphorylation of axin and beta-catenin. FEBS Lett. 458, 247–251.
10. Ding, V. W., Chen, R. H., and McCormick, F. (2000) Differential regulation of glycogensynthase kinase 3beta by insulin and Wnt signaling. J. Biol. Chem. 275, 32475– 32481.
11. Liu, X., Rubin, J. S., and Kimmel, A. R. (2005) Rapid, Wnt-induced changes in GSK3beta associations that regulate betacateninstabilization are mediated by Galpha proteins. Curr. Biol. 15, 1989–1997.
12. Cook, D., Fry, M. J., Hughes, K., Sumathipala,R., Woodgett, J. R., and Dale, T. C. (1996) Wingless inactivates glycogen synthasekinase-3 via an intracellular signalling pathway which involves a protein kinase C. EMBO J. 15, 4526–4536.
13. Sutherland, C., Leighton, I. A., and Cohen,P. (1993) Inactivation of glycogen synthase kinase-3 beta by phosphorylation: new kinase connections in insulin and growthfactorsignalling. Biochem. J. 296 (Pt 1), 15–19.
14. Cole, A., Frame, S., and Cohen, P. (2004)Further evidence that the tyrosine phosphorylationof glycogen synthase kinase-3 (GSK3)in mammalian cells is an autophosphorylationevent. Biochem. J. 377, 249–255.
15. Cole, A. R., Knebel, A., Morrice, N. A., Robertson, L. A., Irving, A. J., Connolly, C. N. et al. (2004) GSK-3 phosphorylation of the Alzheimer epitope within collapsin response mediator proteins regulates axon elongation in primary neurons. J. Biol. Chem. 279, 50176–50180.
16. Hart, M. J., de los, S. R., Albert, I. N.,Rubinfeld, B., and Polakis, P. (1998)Downregulation of beta-catenin by humanAxin and its association with the APC tumor suppressor, beta-catenin and GSK3 beta. Curr. Biol. 8, 573–581.
17. Itoh, K., Krupnik, V. E., and Sokol, S. Y. (1998) Axis determination in Xenopus involves biochemical interactions of axin, glycogen synthase kinase 3 and beta-catenin. Curr. Biol. 8, 591–594.
18. Laemmli, U. K. (1970) Cleavage of structuralproteins during the assembly of thehead of bacteriophage T4. Nature 227,680–685.
19. Bradford, M. M. (1976) A rapid and sensitivemethod for the quantitation of microgramquantities of protein utilizing theprinciple of protein-dye binding. Anal. Biochem. 72, 248–254.
20. Cross, D. A., Culbert, A. A., Chalmers, K. A., Facci, L., Skaper, S. D., and Reith, A. D. (2001) Selective small-molecule inhibitors of glycogen synthase kinase-3 activity protect primary neurones from death. J. Neurochem. 77, 94–102.
21. Cross, D. A., Alessi, D. R., Vandenheede,J. R., McDowell, H. E., Hundal, H. S., and Cohen, P. (1994) The inhibition of glycogen synthase kinase-3 by insulin or insulin-like growth factor 1 in the rat skeletal muscle cell line L6 is blocked by wortmannin, but not by rapamycin: evidence that wortmannin blocks activation of the mitogen-activated protein kinase pathway in L6 cells between Ras and Raf. Biochem. J. 303 (Pt 1), 21–26.
