- Email: [email protected]
High yield expression of non-phosphorylated protein tyrosine kinases in insect cells
Protein Expression and Purification 61 (2008) 204–211
Contents lists available at ScienceDirect
Protein Expression and Purification j o u r n a l h o m e p a g e : w w w . e l s e v i e r. c o m / l o c a t e / y p r e p
High yield expression of non-phosphorylated protein tyrosine kinases in insect cells Leyu Wang *, Meta Foster, Yan Zhang, William R. Tschantz, Lily Yang, Joe Worrall, Christine Loh, Xu Xu Research Technology Center, Pfizer Cambridge Laboratories, 620 Memorial Drive, Cambridge, MA 02139, USA
a r t i c l e
i n f o
Article history: Received 23 April 2008 and in revised form 28 May 2008 Available online 5 June 2008 Keywords: Kinase Phosphatase Phosphorylation Baculovirus Insect cells Coexpression BTK JAK3 EphA2 YopH
a b s t r a c t The key role of kinases in signal transduction and cell growth regulation has been a long standing interest among academics and the pharmaceutical industry. Recombinant enzymes have been used to understand the mechanism of action as well as to screen for chemical inhibitors. The baculo-insect system has been the primary method used to obtain soluble and active kinases, usually producing a mixture of the kinase in various phosphorylation states in different conformations. To obtain a homogenous preparation of non-phosphorylated kinases is critical for biochemical, biophysical and kinetic studies aimed at under standing the mechanism of kinase activation. Taking advantage of the eukaryotic expression property of insect cells, we were able to obtain high yield expression of non-phosphorylated protein tyrosine kinases BTK, JAK3 and Eph2A through coexpression with the tyrosine phosphatase YopH, which suggests that this method can be applied to protein tyrosine kinases in general. We have demonstrated that the fully nonphosphorylated BTK obtained with this method is suitable for various biochemical and kinetic studies. © 2008 Elsevier Inc. All rights reserved.
Introduction Protein kinases play a key role in cellular signaling and cell growth regulation. The research effort dedicated to kinases has increased exponentially over the past two decades due to advances in molecular and cellular biology. It is now well known that there is a close correlation between increased activity of many kinases and a wide range of diseases, especially cancers. For this reason, protein kinases have become a major gene family as therapeutic targets [1,2]. Many kinases are activated by phosphorylation on certain amino acids (Ser, Thr or Tyr) within their activation loops, accompanied by a conformation change [3,4]. Kinase inhibitors have been shown to have preferential affinity to one conformation over the other [5,6]. Isolation of homogenous preparations of phos phorylated and non-phosphorylated versions of a kinase is critical for better understanding ligand–protein interactions. Currently there are two major recombinant approaches to gener ate non-phosphorylated protein kinases, taking advantage of either prokaryotic or eukaryotic expression host cells. One approach is to express kinases in Escherichia coli expression systems. The advan tage of this system is that E. coli cells lack post-translational mod ification; therefore the kinases expressed, in principle, should be in the non-phosphorylated form if the protein kinase has no auto-
* Corresponding author. Fax: +1 617 551 3178. E-mail address: Leyu.Wang@pfizer.com (L. Wang). 1046-5928/$ - see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.pep.2008.05.017
phosphorylation activity (Fig. 1A). In addition, the E. coli expression system has a relatively short gene to protein turn-around time. However, the poor expression and low solubility of human kinases is an intrinsic problem of the E. coli expression system [7]. Some kinases cannot even be directly expressed in E. coli due to their tox icity to the host cells. Efforts have been made to overcome the sol ubility and toxicity issues [8], but previous success is limited and kinase dependent. Additionally, even if soluble kinase is expressed in E. coli, multiple phosphorylated forms may still be observed due to the auto-phosphorylation activity of the kinase [9]. The other alternative approach is to express recombinant kinases in insect cells and treat them with phosphatases during or after purification [6,10–12] (Fig. 1B). Our group had direct successful experience with this approach and produced large amount of non-phosphorylated S6K for structural biology study. The host cells in this expression system are eukaryotic cells and have an excellent track record for producing functional kinases; however the recombinant kinases produced are usually a mixture of the kinase in various phosphory lated forms, as results of post-translational modification from host cells and auto-phosphorylation activity of the kinases themselves. Phosphatase treatment can be used during or after purification to generate non-phosphorylated kinases. Unfortunately, the process is complex and requires subsequent purification steps, and often results in incomplete dephosphorylation or low yield due to protein aggregation. A systematic method to generate fully non-phosphor ylated kinases is needed for biochemical, biophysical and kinetic studies to gain a better understanding of kinase regulation.
L. Wang et al. / Protein Expression and Purification 61 (2008) 204–211
205
Chaperone expression plasmid set, including pG-KJE8 (chaper one expressed dnaK-dnaJ-grpE-groES-groEL), pGro7 (chaperone expressed groES-GroEL), pKJE7 (chaperone expressed dnaK-dnaJgrpE), pG-Tf2 (chaperone expressed groES-groEL-tig) and pTf16 (chaperone expressed tig), was purchased from TaKaRa (Japan). For coexpression, YopH with the same ORF as insect expression sequence was constructed into pCDF-Duet-1 Vector (Novagen). Wild-type BTK and codon optimized BTK for E. coli system (nucleic sequence undisclosed) expressing the exact same amino acid sequence as insect expression described above were cloned into a pET vector (Novagen). Expression of BTK in E. coli cells and solubility assessment
Fig. 1. Comparison of strategies to obtain purified non-phosphorylated kinases. (A) Express and purify kinases directly from E. coli cells with or without phosphatase coexpression. (B) Treat insect cell expressed kinases with phosphatases. (C) Express and purify kinases directly from insect cells with phosphatase coexpression.
Recently, the expression of 62 human kinases was compared side-by-side in E. coli and insect cells, showing protein tyrosine kinases (PTKs) to be the group of kinases most difficult to express in soluble form in E. coli [7]. The tyrosine specific phosphatase YopH [13,14] was shown to be able to increase the solubility of protein tyrosine kinases c-Abl and c-Src in E. coli cells [15]. In the studies described in this paper, we have demonstrated that we can produce soluble non-phosphorylated protein tyrosine kinases (npPTKs) in a systematic approach by coexpressing PTKs with YopH in insect cells (Fig. 1C). Our results suggest that using this method, we could achieve high yield npPTK expression of several human tyrosine kinases, including non-receptor kinase BTK [16], JAK3 [17], as well as receptor kinase EphA2 [18]. We propose that this approach could be used to produce soluble npPTKs in general. Materials and methods Materials General chemicals, including glycerol, BSA, TCEP, DMSO, and reduced glutathione were purchased from Sigma–Aldrich. Tris buffer (1 M) at different pH and 5 M NaCl stock solution were pur chased from BrandNu Laboratroy. Ni–NTA column HisTrap HP (Cat No. 17-5248-02), SP column HiTrap SP HP (Cat No. 17-1152-01) and Heparin column HiTrap Heparin HP (Cat No. 17-0407-01) were pur chased from GE Lifescience. Ultra pure imidazole was purchased from EMD Bioscience. Cell culture materials, antibiotic–antimy cotic solution, 10 mg/ml gentamicin solution, FBS and insect cell growth medium SF-900-II SFM, were purchased from Invitrogen. Antibodies were purchased from Cell Signaling (anti-pTyr), Abcam (Anti-pY223 BTK), Pharmingen (Anti-pY551 BTK), and Promega (anti-His6 and AP-conjugated secondary antibody). Constructs and baculovirus generation Baculoviruses were generated using the Bac-to-Bac system (Invitrogen). The tyrosine phosphatase YopH (NP-052424, full length) from Yersinia was engineered to express the full length pro tein with N-terminal Flag tag for detection purpose. Kinases used as examples were full length BTK (NP_000052, full length) with Cterminal His-tag, JAK3 catalytic domain (NP_000206, amino acids 813–1104) and EphA2 catalytic domain with N-terminal GST tags (NP_004422, amino acids 560–976).
1 Abbreviations used: PTKs, protein tyrosine kinases; npPTKs, non-phosphorylated protein tyrosine kinases; TIPS, Titerless Infected-cells Preservation and Scale-up.
Bacteria strain BL21-AI (Invitrogen) was used in all experiments. Overnight cultured cells were diluted 1:50 into fresh LB medium with antibiotics and grown at 37 °C until OD600 to 0.6 in shaking flasks. Then expression was induced by adding 0.5 mM IPTG and 0.2% l-arabinose (additional 10 ng/ml tetracyclin hydrochloride for cells with pG-KJE8 and pG-tf2) for 4 h. Cells were collected with centrifugation. Cells from 1 ml culture were suspended in 0.5 ml buffer contain ing 50 mM Tris pH 7.6, 150 mM NaCl and 5 mM TCEP, lysed using a sonicator (VirSonic 100, VirTis) with 10 pulses 1 s/pulse. The insol uble protein and cell debris were sedimented through a 30 min centrifugation at 14,000 rpm in a desktop centrifuge machine (Cen trifuge 5417C, Eppendorf) at 4 °C. The supernatants were used as soluble fractions. Preparation of baculovirus infected insect cells (BIIC) stocks The Titerless Infected-cells Preservation and Scale-up (TIPS) method [19] was used for protein expression and coexpression. The key feature of the TIPS method is the preparation of baculo virus infected insect cells (BIIC). Briefly, Sf9 cells were grown in SF-900-II SFM containing 10% FBS, 10 mg/L gentamicin and 10 ml/ L antibiotic-antimycotic. Cells were infected with baculoviruses when the cell density was »0.8 £ 106 cells/ml. The growth curve of infected cells was monitored with a Cedex (Innovatis). Cells were harvested with centrifugation at 1000g for 10 min once the diame ter of the cells had increased by 3–4 lm with viability greater than 90%. Then the cell pellets were resuspended to 1 £ 107 cells/ml in cryopreservation media containing 90% (v/v) SF-900-II SFM, 10% (v/v) DMSO and 1% (w/v) BSA. Protein expression and purification in insect cells Kinases were expressed in Sf21 cells with and without the phos phatase YopH. Sf21 cells were grown in suspension at 27 °C in SF900 SFM II. At a density of 0.8 £ 106 cells/ml, cells were infected with kinases alone or in combination with YopH. 1 BIIC was used to infect 5 L of cells. In the case of coexpression, 1 BIIC of kinase combined with 1BIIC of YopH was used to infect 10 L of cells. The cells were monitored daily with respect to their density, viability, aggregation status and size. Cells were harvested with centrifuga tion at 1000g for 10 min when the viability of the cells was »90% and their size had increased significantly, typically occurring 72 h after BIIC inoculation. His-tagged BTK was purified using metal affinity chromatogra phy followed by ion-exchange chromatography. Cells (1 L) were suspended in 100 ml lysis buffer (Buffer NA) containing 50 mM Tris pH 7.6, 150 mM NaCl, 5 mM TCEP and protease inhibitors. The cells were lysed with a microfluidizer (Watts Fluidair). The insoluble protein and cell debris were sedimented through a 2 h centrifuga tion at 40,000g at 4 °C. The supernatant was filtered and loaded with 5 mM imidazole onto a HisTrap HP column equilibrated with
206
L. Wang et al. / Protein Expression and Purification 61 (2008) 204–211
0.5% buffer NB (buffer NA containng 1 M imidazole). The column was step-washed with 0.5, 2.5, 5 and 10% buffer NB before the pro tein was eluted using 25% buffer NB. The 25% buffer elution was dialyzed against dialysis buffer (Buffer SPA): 50 mM Tris pH 7.6, 150 mM NaCl, 1 mM EDTA, 0.02% Triton X-100 and 2.5 mM DTT over night at 4 °C. The dialyzed protein was loaded onto an ion exchange column HiTrap SP FF, equilibrated with buffer SPA. Proteins were eluted with a linear gradient of 0–80% buffer SPB (buffer SPA con taining 1 M NaCl) and dialyzed against buffer SPA containing 10% glycerol overnight at 4 °C. GST-tagged JAK3 and Eph2A were purified with GST affinity chromatography followed by Heparin chromatography. 1 L of cells was then lysed and centrifuged similarly to the BTK purification in 50 ml lysis buffer GL containing 50 mM Tris pH 7.4, 150 mM NaCl, 10% glycerol and 5 mM DTT. The supernatant was rocked with 5 ml Glutathione FF resin overnight at 4 °C. The resin was washed with buffer GL and buffer HA containing 50 mM Tris pH 7.4 and 2 mM DTT before the proteins were eluted using 50 mM Tris pH 8.0, 2 mM DTT and 20 mM glutathione (reduced). Then the eluted protein was loaded onto a Heparin HP column equilibrated with buffer HA. Pro teins were eluted with a linear gradient 0–100% buffer HB (HA con taining 0.75 M NaCl). Elution peaks were analyzed with SDS–PAGE, and dialyzed against storage buffer containing 50 mM Tris pH 8.0, 50 mM NaCl, 1 mM EGTA, 1 mM DTT and 25% glycerol. Concentration of the proteins was determined with Bradford assay (Bio-Rad). All proteins were flash-frozen in liquid nitrogen following final dialysis and stored at ¡80 °C. BTK auto-phosphorylation and phosphorylation with Syk BTK auto-phosphorylation and phosphorylation by Syk (Invitro gen) were monitored with Western blots and mass spectroscopy
analysis. Purified BTK was diluted to the final concentration of 0.1 mg/ml in 50 mM HEPES pH 7.5, 0.01% BRIJ-35, 10 mM MgCl2, 1 mM EGTA with and without Syk (mole ratio of BTK and Syk is 20:1), and the reactions were initiated by adding 0.5 mM ATP at room temperature. Samples were taken out as the reaction pro ceeded, and the reactions were stopped with EDTA to a final con centration of 100 mM. The samples were analyzed by SDS–PAGE/ Western bloting and submitted for mass spectroscopy analysis. Mass spectroscopy The protein mass was determined on a ThermoFinnigan LTQ mass spectrometer coupled with an Agilent Technologies 1100 HPLC system. Buffer A (water with 0.2% formic acid) and buffer B (actonitrile with 0.2% formic acid) were used as mobile phases for the HPLC systems. Protein samples were loaded onto a Microtrap protein column (Michrom Bioresources Inc product 0001309) preequilibrated with 2% buffer B with a flow rate of 0.2 ml/min, and eluted with a gradient scheme: a linear increase to 70% buffer B in 8 min, then another linear increase to 98% buffer B in 2 min. The MS data were collected in positive ion mode by full MS scan from m/z 300 to 2000. The protein molecular weight was calculated by deconvoluting the mass spectrum with ProMass software. SDS–PAGE and Western blot All SDS–PAGE analysis was performed using NuPAGE 4–12% Bis–Tris Gels with the MES running buffer system (Invitrogen). Gels were either stained with Coomassie Blue or transblotted to PVDF membrane using the iBlot system (Invitrogen) for Western analysis. The Western blot was developed with Promega Western Blue method. Briefly, the PVDF membrane was incubated with the
Fig. 2. Expression strategies to improve BTK solubility and yield in E. coli cells. (A) Coexpression of chaperones affects soluble BTK level. Soluble fractions and total lysates are labeled as S and T, respectively. Samples loaded are equivalent to 20 ll culture. BTK alone (Lane 1) and coexpressed with chaperone groES-groEL-tig (Lane 2), Tig (Lane 3), groES-groEL (Lane 4), dnaK-dnaJ-grpE-groES-groEL (Lane 5) and dnaK-dnaJ-grpE (Lane 6). (B) Coexpression of YopH improves soluble BTK level by about onefold. Lane 1: soluble fraction from 2.5 ll BTK expressing sf21 cells; lane 2: soluble fraction from 20 ll E. coli culture with BTK expression alone; lane 3: soluble fraction from 20 ll E. coli culture with BTK and YopH coexpression. (C) Expression of BTK without codon optimization. Lane 1: total lysate from 20 ll wildt-type BTK expressing E. coli culture; lane 2: soluble fraction from 20 ll wild-type BTK expressing E. coli culture. (D) Expression of one codon optimized BTK. Lane 1: total lysate from 20 ll a codon optimized BTK expressing E. coli culture; lane 2: soluble fraction from 20 ll the codon optimized BTK expressing E. coli culture. (A and B) Western blotting results with anti-His6 antibody, (C and D) Coomassie Blue stained SDS–PAGE.
L. Wang et al. / Protein Expression and Purification 61 (2008) 204–211
primary antibody (all primary antibodies in this work were used in 1:1000 dilution in 5% Carnation nonfat milk in TBST buffer) after it was blotted with 5% Carnation nonfat milk in TBST buffer for 30 min. The membrane was washed three times with TBST (10 min/ wash). Then the membrane was probed with Promega AP-conju gated secondary antibody in TBST for 30 min. Then the membrane was developed with 5 ml Western Blue Stabilized Substrate for Alkaline Phosphatase (Promega) after being washed three times with TBST (10 min/wash). Results and discussion We constructed the idea to develop a method to obtain npPTKs while studying the activation mechanism of BTK in vitro. BTK is one of many PTKs shown to be activated by its tyrosine phosphoryla tion [20–22]. To understand the role of phosphorylation in human BTK kinase activation and its impact on small molecule inhibi tor discovery, we needed to develop expression and purification approaches which would allow us to obtain significant amounts of human full length BTK protein in the fully non-phosphorylated state (npBTK). An E. coli expression system was tried for its known lack of post-translational modification and literature precedence [23]. Unfortunately, we only obtained a low expression level in E. coli using the wild-type human BTK gene sequence, the major ity of which accumulated in inclusion bodies. Coexpressing with chaperone groES-groEL had increased soluble BTK in the lysate (Fig. 2A), but had no significant impact on the BTK yield after puri fication due to extra steps to eliminate co-purified chaperones. To address the problem of solubility, we attempted phosphatase coex
207
pression. Tyrosine phosphatase YopH has been shown to increase the solubility of protein tyrosine kinases Abl and Src in E. coli [15]. Using a similar method, we found that YOP coexpression only mar ginally increased BTK solubility with a low yield not sustainable for biochemical or structural biology studies (Fig. 2B). We were able to increase the BTK expression level more than tenfold with gene optimization (Fig. 2C and D); however the soluble BTK level remained low. Insect cell expression and post-expression phosphatase treat ment approaches were explored in parallel (Fig. 1B). As eukaryotic host cells, insect cells have been shown to be better host cells for human protein expression [7,24]. Baculovirus encoding hexa-His tine tagged full length human BTK was produced and infection of sf21 insect cells was performed using the Titerless Infected-cells Preservation and Scale-up (TIPS) method described previously [19]. Protein expression analysis suggested that human BTK protein expressed well in insect cells and the majority of the protein was in the soluble fraction. After metal affinity purification followed by ion-exchange chromatography (see Materials and methods for detail), we were able to obtain »4 mg BTK protein with more than 90% purity from 1 L sf21 cells at a density of 1.2 £ 106 cells/ml (Fig. 3A). When examined by Western analysis using pY antibody and mass spectroscopy, BTK obtained this way was found to be in a mix ture of highly heterogeneous phosphorylated states (Fig. 3B and D). We next tried to generate npBTK by treating purified BTK from the insect cells with a variety of phosphatases (k-PPase, YOP, TC PTP and CIP from New England BioLabs) followed by repurification. This approach turned out to be labor intensive and inefficient, with BTK often precipitating out of the solution.
Fig. 3. BTK from insect cells with or without coexpression with YopH. (A) Equal amount of purified BTK was analyzed on SDS–PAGE, stained with Coomassie Blue. Lane 1: 5 lg purified BTK without YopH coexpression; lane 2: 5 lg purified BTK with YopH coexpression. (B) Equal amount of purified BTK was analyzed on SDS–PAGE and Western-blot ted with pY antibody from Cell Signaling. Lane 1: 200 ng purified BTK without YopH coexpression; lane 2: 200 ng purified BTK with YopH coexpression. (C) Mass Spectrum analysis of purified BTK from insect cells with YopH coexpression. (D) Mass spectrum analysis of purified BTK from insect cells. The theoretical molecular weight of this construct is 81, 244 Da. BTK from insect cells without YopH coexpression is a mixture of protein at various phosphorylation stages with an average of two phosphorylation sites. “P” stands for one phosphorylation modification.
208
L. Wang et al. / Protein Expression and Purification 61 (2008) 204–211
Given the poor results of the post-expression treatment with phosphatases and the E. coli expression, we explored the coex pression of BTK with the tyrosine phosphatase YopH in insect cells for the following reasons: (1) insect cells are better hosts to express soluble kinases compared to E. coli cells [7]; (2) BTK is a protein tyrosine kinase and is known to be activated by its tyro sine phosphorylation (Y551 and Y223); (3) YopH is a phospha tase highly specific for phosphotyrosine [14]. To test this idea, we generated baculovirus encoding full length YopH with an Nterminal FLAG tag for coexpression. With our coexpression and purification procedure (see Materials and methods for detail), we routinely obtained »4 mg protein with >90% purity from 1 L sf21 cells (Fig. 3A), comparable to BTK yield without YopH coex pression. Data from Western-blot and mass spectrum analysis suggested that tyrosine residues in BTK from the insect cells with YopH coexpression were not phosphorylated (Fig. 3B and C). The observed molecular weight of the BTK protein from mass spectroscopy corresponded to the fully non-phosphorylated
full length BTK protein. The recombinant npBTK produced this way was shown to retain its biological function. It can undergo auto-phosphorylation reaction on multiple residues upon incu bation with ATP and Mg2+ in a time dependent fashion (Fig. 4A). Additionally the BTK upstream kinase Syk can further accelerate phosphorylation modification on the recombinant BTK (Fig. 4B). We have also examined the phosphorylation status on Y223 and Y551 residues, phosphorylation of which has been suggested to be essential for BTK to be fully active [20–22]. Western analy sis of the BTK auto-phosphorylation reaction using pY551 and pY223 specific antibodies suggested that both tyrosine resi dues can be phosphorylated in a time dependent manner and the phosphorylation on the Y551 residue appeared faster than the phosphorylation on Y223 residue (Fig. 4C and D). Using this approach and combined with site-directed mutagenesis, we have performed various kinetic and mechanism studies to explore the role of Y223 and Y551 phosphorylation in BTK activity regulation (results will be described in a separate manuscript).
Fig. 4. Activation analysis of purified npBTK. (A) Mass Spectrum analysis on the auto-phosphorylation of npBTK. npBTK was incubated with ATP/Mg2+ at room temperature, and samples were withdrawn at certain time points and analyzed with a mass spectrometer. (B) Mass Spectrum analysis on the phosphorylation of npBTK by upstream kinase Syk. npBTK was incubated with Syk and ATP/Mg2+ at room temperature, and samples were withdrawn at certain time points and analyzed with a mass spectrometer. (C and D) Western analysis on npBTK auto-phosphorylation with BTK pY551 and pY223 antibodies, respectively. Lanes 1, 2, 3, 4 and 5 are reactions at 0, 5, 10, 15 and 30 min, respectively. BTK (200 ng) was loaded in each lane. The details of the experiment are described in Materials and methods.
L. Wang et al. / Protein Expression and Purification 61 (2008) 204–211
PTKs have been shown to be the most difficult kinases to express in soluble form using E. coli [7]. Given our success with producing BTK using YopH insect cell coexpression method, we decided to test our insect cell YopH coexpression approach with other PTKs. We applied the method on additional human PTKs including cyto plasmic kinase JAK3 and receptor kinase EphA2. What is signifi cant is that this method worked equally well on these additional unrelated PTKs. The JAK3 kinase domain (JH1 domain, amino acid 813–1104) and the EphA2 intracellular catalytic domain (amino acid 560–976) were GST-tagged. Both proteins were purified with GST affinity chromatography followed by Heparin chromatography (see Materials and methods for detail), and analyzed with Western and mass spectroscopy. As shown in Fig. 5, expression of JAK3 alone resulted in heterogeneous phosphorylation on the protein (Fig. 5A and D), while coexpression with YopH phosphatase pro duced a completely de-phosphorylated JAK3 (Fig. 5B and C). The molecular weight of the major peak agreed well with the calcu lated non-phosphorylated JAK3 with N-Met removal, the molecu lar weight of the minor peak was equal to the theoretical molecu lar weight of acetylated JAK3. Similar results were observed with EphA2 recombinant protein (Fig. 6). There were two more minor peaks (MW 74, 822 and 74, 906) other than the major non-phos phorylated EphA2 peak (MW 74, 739) in the mass spectroscopy (Fig. 6C) of YopH coexpression product. Further studies are needed to understand whether these are Ser/Thr phosphorylation states of the EphA2 protein when expressed in insect cells. Compared with the enzymatic post-expression treatment (Fig. 1B), the baculovirus mediated coexpression (Fig. 1C) has several
209
advantages besides cost. First, it greatly simplifies the overall pro cess which is evident in Fig. 1, and does not require enzymatic treatment and re-purification. Second, post-expression enzymatic treatment is an optimization process like any assay development. It needs a lot of work to identify the optimal condition. Third, post-expression enzymatic treatment requires prolong incubation of purified protein on ice or room temperature. The protein stabil ity will limit this approach. Fourth, sometimes there is no easy translation from small scale test to scale-up for post-expression treatment. This actually was the bottle neck when we treated BTK with phosphatases. With the baculovirus mediated coexpression, certain phosphatases might also affect the expression and folding ability of insect cells therefore causing some undesired results. Fortunately it is not the case for YopH phosphatase. With all PTKs tested, the yield for npPTKs stays the same if not better. An alternative option to coexpression is to use virus generated from a single plasmid carrying both PTKs and YopH using duel vec tors (pFastBac DUAL etc). In principle, it should produce npPTKs. The downsides are (1) it only produces npPTKs; (2) for every PTK, one dual expression construct needs to be made; (3) one will lose the flexibility to adjust kinase-phosphatase ratio necessary to achieve optimal expression condition. On the other hand, our strategy will provide the flexibility to adjust kinase:phosphatase ratio to achieve optimal npPTK yield, to produce PTK or npPTK with only one YopH construct. Therefore it provides flexibility as well as cost efficiency. In summary, we report a method to obtain high yield expression of npPTKs from insect cells with YopH coexpression. Coexpression of kinase and phosphatase in mammalian cells has long been used
Fig. 5. JAK3 from insect cells with or without coexpression with YopH. (A) Equal amount of purified JAK3 was analyzed on SDS–PAGE, stained with Coomassie Blue. Lane 1: 5 lg purified JAK3 without YopH coexpression; lane 2: 5 lg purified JAK3 with YopH coexpression. (B) Equal amount of purified JAK3 was analyzed on SDS–PAGE and Western-blotted with pY antibody from Cell Signaling. Lane 1: 200 ng purified JAK3 without YopH coexpression; lane 2: 200 ng purified JAK3 with YopH coexpression. (C) Mass Spectrum analysis of purified JAK3 with YopH coexpression. (D) Mass Spectrum analysis of purified JAK3 from insect cells without YopH coexpression. The theoretical molecular weight of this construct is 59, 642 Da or 59, 511 Da (N-terminal Met removal).
210
L. Wang et al. / Protein Expression and Purification 61 (2008) 204–211
Fig. 6. EphA2 from insect cells with or without coexpression with YopH. (A) Equal amount of purified EphA2 was analyzed on SDS–PAGE, stained with Coomassie. Lane 1: 5 lg purified EphA2 without YopH coexpression; lane 2: 5 lg purified EphA2 with YopH coexpression. (B) Equal amount of purified EphA2 was analyzed on SDS–PAGE and Western-blotted with pY antibody from Cell Signaling. Lane 1: 200 ng purified EphA2 without YopH coexpression; lane 2: 200 ng purified EphA2 with YopH coexpression. (C) Mass Spectrum analysis of purified EphA2 from insect cells with YopH coexpression. (D) Mass Spectrum analysis of purified EphA2 from insect cells without YopH coex pression. The theoretical molecular weight of this construct is 74, 906 Da. The molecular weight of unphosphorylated peaks (74, 739 Da in C and 74, 742 Da in D) »165 Da less than the theoretical molecular weight may be due to other modifications and have no simple explanation.
to study the kinase activation mechanism and elucidate its biolog ical function. It has also been used in E. coli expression systems to generate non-phosphorylated kinases or improve kinase yield [8,15,25,26]. To our knowledge, this is the first report to employ kinase/phosphatase coexpression strategy from insect cells to produce npPTKs. Taking advantage of the insect cell’s eukaryotic expression system, which has an excellent track record for gener ating human kinases, we were able to produce functional and sol uble npPTKs through coexpression with the tyrosine phosphatase YopH. This method was tested against cytoplasmic tyrosine kinases as well as receptor tyrosine kinase, and achieved equally positive results. These results suggest that the insect cell YopH coexpression method can be generally applied to other tyrosine kinases. It greatly simplifies the process to generate npPTKs, and will contribute to bet ter understanding the mechanism of the tyrosine kinase activation process and the discovery of kinase inhibitors with novel inhibition mechanisms. It is reasonable to believe that a similar process can be applied to generate non-phosphorylated Ser/Thr kinases in insect cells by coexpressing them with a Ser/Thr phosphatase. Acknowledgments We thank Alexis C. Meng for her help in preparing of the manu script, Robert Stanton for his helpful inputs. References [1] M.E. Noble, J.A. Endicott, L.N. Johnson, Protein kinase inhibitors: insights into drug design from structure, Science 303 (2004) 1800–1805.
[2] J.S. Sebolt-Leopold, J.M. English, Mechanisms of drug inhibition of signaling molecules, Nature 441 (2006) 457–462. [3] S.R. Hubbard, L. Wei, L. Ellis, W.A. Hendrickson, Crystal structure of the tyrosine kinase domain of the human insulin receptor, Nature 372 (1994) 746–754. [4] S.R. Hubbard, Crystal structure of the activated insulin receptor tyrosine kinase in complex with peptide substrate and ATP analog, EMBO J. 16 (1997) 5572–5581. [5] S.R. Hubbard, Protein tyrosine kinases: autoregulation and small-molecule inhibition, Curr. Opin. Struct. Biol. 12 (2002) 735–741. [6] T. Schindler, W. Bornmann, P. Pellicena, W.T. Miller, B. Clarkson, J. Kuriyan, Structural mechanism for STI-571 inhibition of Abelson tyrosine kinase, Sci ence 289 (2000) 1938–1942. [7] S.P. Chambers, D.A. Austen, J.R. Fulghum, W.M. Kim, High-throughput screen ing for soluble recombinant expressed kinases in Escherichia coli and insect cells, Protein Expr. Purif. 36 (2004) 40–47. [8] R.A. Elling, B.T. Tangonan, D.M. Penny, J.T. Smith, D.E. Vincent, S.K. Hansen, T. O’Brien, M.J. Romanowski, Mouse Aurora A: expression in Escherichia coli and purification, Protein Expr. Purif. 54 (2007) 139–146. [9] P. Du, P. Loulakis, C. Luo, A. Mistry, S.P. Simons, P.K. LeMotte, F. Rajamohan, K. Rafidi, K.G. Coleman, K.F. Geoghegan, Z. Xie, Phosphorylation of serine res idues in histidine-tag sequences attached to recombinant protein kinases: a cause of heterogeneity in mass and complications in function, Protein Expr. Purif. 44 (2005) 121–129. [10] C.K. Smith, D. Carr, T.W. Mayhood, W. Jin, K. Gray, T.W. Windsor, Expression and purification of phosphorylated and non-phosphorylated human MEK1, Protein Expr. Purif. 52 (2007) 446–456. [11] P.M. Chan, S. Ilangumaran, J. La Rose, A. Chakrabartty, R. Rottapel, Autoinhi bition of the Kit receptor tyrosine kinase by the cytosolic juxtamembrane region, Mol. Cell. Biol. 23 (2003) 3067–3078. [12] J.H. Till, M. Becerra, A. Watty, Y. Lu, Y. Ma, T.A. Neubert, S.J. Burden, S.R. Hub bard, Crystal structure of the MuSK tyrosine kinase insights into receptor auto regulation, Structure 10 (2002) 1187–1196. [13] K.L. Guan, J.E. Dixon, Protein tyrosine phosphatase activity of an essential viru lence determinant in Yersinia, Science 249 (1990) 553–556. [14] Z.Y. Zhang, J.C. Clemens, H.L. Schubert, J.A. Stuckey, M.W. Fischer, D.M. Hume, M.A. Saper, J.E. Dixon, Expression, purification, and physicochemical character ization of a recombinant Yersinia protein tyrosine phosphatase, J. Biol. Chem. 267 (1992) 23759–23766.
L. Wang et al. / Protein Expression and Purification 61 (2008) 204–211
[15] M.A. Seeliger, M. Young, M.N. Henderson, P. Pellicena, D.S. King, A.M. Falick, J. Kuriyan, High yield bacterial expression of active c-Abl and c-Src tyrosine kinases, Protein Sci. 14 (2005) 3135–3139. [16] C.I. Smith, B. Baskin, P. Humire-Greiff, J.N. Zhou, P.G. Olsson, H.S. Maniar, P. Kjellen, J.D. Lambris, B. Christensson, L. Hammarstrom, Expression of Bruton’s agammaglobulinemia tyrosine kinase gene, BTK, is selectively down-regulated in T lymphocytes and plasma cells, J. Immuol. 152 (1994) 557–565. [17] S.G. Rane, E.P. Reddy, JAK3: a novel JAK kinase associated with terminal differ entiation of hematopoietic cells, Oncogene 9 (1994) 2415–2423. [18] R.A. Lindberg, T. Hunter, cDNA cloning and characterization of eck, an epi thelial cell receptor protein-tyrosine kinase in the eph/elk family of protein kinase, Mol. Cell. Biol. 10 (1990) 6316–6324. [19] D.J. Wasilko, E. Lee, TIPS: titerless infected-cells preservation and scale-up, Bio Processing J. 5 (3) (2006) 29–32. [20] D.E. Afar, H. Park, B.W. Howell, D.J. Rawlings, J. Cooper, O.N. Witte, Regulation of Btk by Src family tyrosine kinases, Mol. Cell. Biol. 16 (1996) 3465–3471.
211
[21] H. Park, M.I. Wahl, D.E. Afar, C.W. Turck, D.J. Rawlings, C. Tam, A.M. Scharen berg, J.P. Kinet, O.N. Witt, Regulation of BTK function by a major phosphoryla tion site within the SH3 domain, Immunity 4 (1996) 515–525. [22] D.J. Rawlings, A.M. Scharenberg, H. Park, M.I. Wahl, S. Lin, R.M. Kato, A.C. Fluck iger, O.N. Witte, J.P. Kinet, Activation of BTK by a phosphorylation mechanism initiated by SRC family kinases, Science 291 (1996) 822–825. [23] K. Bence, W. Ma, T. Kozasa, X.Y. Huang, Direct stimulation of Bruton’s tyrosine kinase by Gq-protein a-subunit, Nature 389 (1997) 296–299. [24] J.M. Vlak, R.J. Keus, Baculovirus expression vector system for production of viral vaccines, Adv. Biotechnol. Process. 14 (1990) 91–128. [25] T. Matsui, K. Tanihara, T. Date, Expression of unphosphorylated form of human double-stranded RNA-activated protein kinase in Escherichia coli, Biochem. Bio phys. Res. Commun. 284 (2001) 798–807. [26] W. Wang, A. Marimuthu, J. Tsai, A. Kumar, H.I. Krupka, C. Zhang, B. Powell, Y. Suzuki, H. Nguyen, M. Tabrizizad, C. Luu, B.L. West, Structural characterization of autoinhibited c-Met kinase produced by coexpression in bacteria with phos phatase, Proc. Natl. Acad. Sci. USA 103 (2006) 3563–3568.
Contents lists available at ScienceDirect
Protein Expression and Purification j o u r n a l h o m e p a g e : w w w . e l s e v i e r. c o m / l o c a t e / y p r e p
High yield expression of non-phosphorylated protein tyrosine kinases in insect cells Leyu Wang *, Meta Foster, Yan Zhang, William R. Tschantz, Lily Yang, Joe Worrall, Christine Loh, Xu Xu Research Technology Center, Pfizer Cambridge Laboratories, 620 Memorial Drive, Cambridge, MA 02139, USA
a r t i c l e
i n f o
Article history: Received 23 April 2008 and in revised form 28 May 2008 Available online 5 June 2008 Keywords: Kinase Phosphatase Phosphorylation Baculovirus Insect cells Coexpression BTK JAK3 EphA2 YopH
a b s t r a c t The key role of kinases in signal transduction and cell growth regulation has been a long standing interest among academics and the pharmaceutical industry. Recombinant enzymes have been used to understand the mechanism of action as well as to screen for chemical inhibitors. The baculo-insect system has been the primary method used to obtain soluble and active kinases, usually producing a mixture of the kinase in various phosphorylation states in different conformations. To obtain a homogenous preparation of non-phosphorylated kinases is critical for biochemical, biophysical and kinetic studies aimed at under standing the mechanism of kinase activation. Taking advantage of the eukaryotic expression property of insect cells, we were able to obtain high yield expression of non-phosphorylated protein tyrosine kinases BTK, JAK3 and Eph2A through coexpression with the tyrosine phosphatase YopH, which suggests that this method can be applied to protein tyrosine kinases in general. We have demonstrated that the fully nonphosphorylated BTK obtained with this method is suitable for various biochemical and kinetic studies. © 2008 Elsevier Inc. All rights reserved.
Introduction Protein kinases play a key role in cellular signaling and cell growth regulation. The research effort dedicated to kinases has increased exponentially over the past two decades due to advances in molecular and cellular biology. It is now well known that there is a close correlation between increased activity of many kinases and a wide range of diseases, especially cancers. For this reason, protein kinases have become a major gene family as therapeutic targets [1,2]. Many kinases are activated by phosphorylation on certain amino acids (Ser, Thr or Tyr) within their activation loops, accompanied by a conformation change [3,4]. Kinase inhibitors have been shown to have preferential affinity to one conformation over the other [5,6]. Isolation of homogenous preparations of phos phorylated and non-phosphorylated versions of a kinase is critical for better understanding ligand–protein interactions. Currently there are two major recombinant approaches to gener ate non-phosphorylated protein kinases, taking advantage of either prokaryotic or eukaryotic expression host cells. One approach is to express kinases in Escherichia coli expression systems. The advan tage of this system is that E. coli cells lack post-translational mod ification; therefore the kinases expressed, in principle, should be in the non-phosphorylated form if the protein kinase has no auto-
* Corresponding author. Fax: +1 617 551 3178. E-mail address: Leyu.Wang@pfizer.com (L. Wang). 1046-5928/$ - see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.pep.2008.05.017
phosphorylation activity (Fig. 1A). In addition, the E. coli expression system has a relatively short gene to protein turn-around time. However, the poor expression and low solubility of human kinases is an intrinsic problem of the E. coli expression system [7]. Some kinases cannot even be directly expressed in E. coli due to their tox icity to the host cells. Efforts have been made to overcome the sol ubility and toxicity issues [8], but previous success is limited and kinase dependent. Additionally, even if soluble kinase is expressed in E. coli, multiple phosphorylated forms may still be observed due to the auto-phosphorylation activity of the kinase [9]. The other alternative approach is to express recombinant kinases in insect cells and treat them with phosphatases during or after purification [6,10–12] (Fig. 1B). Our group had direct successful experience with this approach and produced large amount of non-phosphorylated S6K for structural biology study. The host cells in this expression system are eukaryotic cells and have an excellent track record for producing functional kinases; however the recombinant kinases produced are usually a mixture of the kinase in various phosphory lated forms, as results of post-translational modification from host cells and auto-phosphorylation activity of the kinases themselves. Phosphatase treatment can be used during or after purification to generate non-phosphorylated kinases. Unfortunately, the process is complex and requires subsequent purification steps, and often results in incomplete dephosphorylation or low yield due to protein aggregation. A systematic method to generate fully non-phosphor ylated kinases is needed for biochemical, biophysical and kinetic studies to gain a better understanding of kinase regulation.
L. Wang et al. / Protein Expression and Purification 61 (2008) 204–211
205
Chaperone expression plasmid set, including pG-KJE8 (chaper one expressed dnaK-dnaJ-grpE-groES-groEL), pGro7 (chaperone expressed groES-GroEL), pKJE7 (chaperone expressed dnaK-dnaJgrpE), pG-Tf2 (chaperone expressed groES-groEL-tig) and pTf16 (chaperone expressed tig), was purchased from TaKaRa (Japan). For coexpression, YopH with the same ORF as insect expression sequence was constructed into pCDF-Duet-1 Vector (Novagen). Wild-type BTK and codon optimized BTK for E. coli system (nucleic sequence undisclosed) expressing the exact same amino acid sequence as insect expression described above were cloned into a pET vector (Novagen). Expression of BTK in E. coli cells and solubility assessment
Fig. 1. Comparison of strategies to obtain purified non-phosphorylated kinases. (A) Express and purify kinases directly from E. coli cells with or without phosphatase coexpression. (B) Treat insect cell expressed kinases with phosphatases. (C) Express and purify kinases directly from insect cells with phosphatase coexpression.
Recently, the expression of 62 human kinases was compared side-by-side in E. coli and insect cells, showing protein tyrosine kinases (PTKs) to be the group of kinases most difficult to express in soluble form in E. coli [7]. The tyrosine specific phosphatase YopH [13,14] was shown to be able to increase the solubility of protein tyrosine kinases c-Abl and c-Src in E. coli cells [15]. In the studies described in this paper, we have demonstrated that we can produce soluble non-phosphorylated protein tyrosine kinases (npPTKs) in a systematic approach by coexpressing PTKs with YopH in insect cells (Fig. 1C). Our results suggest that using this method, we could achieve high yield npPTK expression of several human tyrosine kinases, including non-receptor kinase BTK [16], JAK3 [17], as well as receptor kinase EphA2 [18]. We propose that this approach could be used to produce soluble npPTKs in general. Materials and methods Materials General chemicals, including glycerol, BSA, TCEP, DMSO, and reduced glutathione were purchased from Sigma–Aldrich. Tris buffer (1 M) at different pH and 5 M NaCl stock solution were pur chased from BrandNu Laboratroy. Ni–NTA column HisTrap HP (Cat No. 17-5248-02), SP column HiTrap SP HP (Cat No. 17-1152-01) and Heparin column HiTrap Heparin HP (Cat No. 17-0407-01) were pur chased from GE Lifescience. Ultra pure imidazole was purchased from EMD Bioscience. Cell culture materials, antibiotic–antimy cotic solution, 10 mg/ml gentamicin solution, FBS and insect cell growth medium SF-900-II SFM, were purchased from Invitrogen. Antibodies were purchased from Cell Signaling (anti-pTyr), Abcam (Anti-pY223 BTK), Pharmingen (Anti-pY551 BTK), and Promega (anti-His6 and AP-conjugated secondary antibody). Constructs and baculovirus generation Baculoviruses were generated using the Bac-to-Bac system (Invitrogen). The tyrosine phosphatase YopH (NP-052424, full length) from Yersinia was engineered to express the full length pro tein with N-terminal Flag tag for detection purpose. Kinases used as examples were full length BTK (NP_000052, full length) with Cterminal His-tag, JAK3 catalytic domain (NP_000206, amino acids 813–1104) and EphA2 catalytic domain with N-terminal GST tags (NP_004422, amino acids 560–976).
1 Abbreviations used: PTKs, protein tyrosine kinases; npPTKs, non-phosphorylated protein tyrosine kinases; TIPS, Titerless Infected-cells Preservation and Scale-up.
Bacteria strain BL21-AI (Invitrogen) was used in all experiments. Overnight cultured cells were diluted 1:50 into fresh LB medium with antibiotics and grown at 37 °C until OD600 to 0.6 in shaking flasks. Then expression was induced by adding 0.5 mM IPTG and 0.2% l-arabinose (additional 10 ng/ml tetracyclin hydrochloride for cells with pG-KJE8 and pG-tf2) for 4 h. Cells were collected with centrifugation. Cells from 1 ml culture were suspended in 0.5 ml buffer contain ing 50 mM Tris pH 7.6, 150 mM NaCl and 5 mM TCEP, lysed using a sonicator (VirSonic 100, VirTis) with 10 pulses 1 s/pulse. The insol uble protein and cell debris were sedimented through a 30 min centrifugation at 14,000 rpm in a desktop centrifuge machine (Cen trifuge 5417C, Eppendorf) at 4 °C. The supernatants were used as soluble fractions. Preparation of baculovirus infected insect cells (BIIC) stocks The Titerless Infected-cells Preservation and Scale-up (TIPS) method [19] was used for protein expression and coexpression. The key feature of the TIPS method is the preparation of baculo virus infected insect cells (BIIC). Briefly, Sf9 cells were grown in SF-900-II SFM containing 10% FBS, 10 mg/L gentamicin and 10 ml/ L antibiotic-antimycotic. Cells were infected with baculoviruses when the cell density was »0.8 £ 106 cells/ml. The growth curve of infected cells was monitored with a Cedex (Innovatis). Cells were harvested with centrifugation at 1000g for 10 min once the diame ter of the cells had increased by 3–4 lm with viability greater than 90%. Then the cell pellets were resuspended to 1 £ 107 cells/ml in cryopreservation media containing 90% (v/v) SF-900-II SFM, 10% (v/v) DMSO and 1% (w/v) BSA. Protein expression and purification in insect cells Kinases were expressed in Sf21 cells with and without the phos phatase YopH. Sf21 cells were grown in suspension at 27 °C in SF900 SFM II. At a density of 0.8 £ 106 cells/ml, cells were infected with kinases alone or in combination with YopH. 1 BIIC was used to infect 5 L of cells. In the case of coexpression, 1 BIIC of kinase combined with 1BIIC of YopH was used to infect 10 L of cells. The cells were monitored daily with respect to their density, viability, aggregation status and size. Cells were harvested with centrifuga tion at 1000g for 10 min when the viability of the cells was »90% and their size had increased significantly, typically occurring 72 h after BIIC inoculation. His-tagged BTK was purified using metal affinity chromatogra phy followed by ion-exchange chromatography. Cells (1 L) were suspended in 100 ml lysis buffer (Buffer NA) containing 50 mM Tris pH 7.6, 150 mM NaCl, 5 mM TCEP and protease inhibitors. The cells were lysed with a microfluidizer (Watts Fluidair). The insoluble protein and cell debris were sedimented through a 2 h centrifuga tion at 40,000g at 4 °C. The supernatant was filtered and loaded with 5 mM imidazole onto a HisTrap HP column equilibrated with
206
L. Wang et al. / Protein Expression and Purification 61 (2008) 204–211
0.5% buffer NB (buffer NA containng 1 M imidazole). The column was step-washed with 0.5, 2.5, 5 and 10% buffer NB before the pro tein was eluted using 25% buffer NB. The 25% buffer elution was dialyzed against dialysis buffer (Buffer SPA): 50 mM Tris pH 7.6, 150 mM NaCl, 1 mM EDTA, 0.02% Triton X-100 and 2.5 mM DTT over night at 4 °C. The dialyzed protein was loaded onto an ion exchange column HiTrap SP FF, equilibrated with buffer SPA. Proteins were eluted with a linear gradient of 0–80% buffer SPB (buffer SPA con taining 1 M NaCl) and dialyzed against buffer SPA containing 10% glycerol overnight at 4 °C. GST-tagged JAK3 and Eph2A were purified with GST affinity chromatography followed by Heparin chromatography. 1 L of cells was then lysed and centrifuged similarly to the BTK purification in 50 ml lysis buffer GL containing 50 mM Tris pH 7.4, 150 mM NaCl, 10% glycerol and 5 mM DTT. The supernatant was rocked with 5 ml Glutathione FF resin overnight at 4 °C. The resin was washed with buffer GL and buffer HA containing 50 mM Tris pH 7.4 and 2 mM DTT before the proteins were eluted using 50 mM Tris pH 8.0, 2 mM DTT and 20 mM glutathione (reduced). Then the eluted protein was loaded onto a Heparin HP column equilibrated with buffer HA. Pro teins were eluted with a linear gradient 0–100% buffer HB (HA con taining 0.75 M NaCl). Elution peaks were analyzed with SDS–PAGE, and dialyzed against storage buffer containing 50 mM Tris pH 8.0, 50 mM NaCl, 1 mM EGTA, 1 mM DTT and 25% glycerol. Concentration of the proteins was determined with Bradford assay (Bio-Rad). All proteins were flash-frozen in liquid nitrogen following final dialysis and stored at ¡80 °C. BTK auto-phosphorylation and phosphorylation with Syk BTK auto-phosphorylation and phosphorylation by Syk (Invitro gen) were monitored with Western blots and mass spectroscopy
analysis. Purified BTK was diluted to the final concentration of 0.1 mg/ml in 50 mM HEPES pH 7.5, 0.01% BRIJ-35, 10 mM MgCl2, 1 mM EGTA with and without Syk (mole ratio of BTK and Syk is 20:1), and the reactions were initiated by adding 0.5 mM ATP at room temperature. Samples were taken out as the reaction pro ceeded, and the reactions were stopped with EDTA to a final con centration of 100 mM. The samples were analyzed by SDS–PAGE/ Western bloting and submitted for mass spectroscopy analysis. Mass spectroscopy The protein mass was determined on a ThermoFinnigan LTQ mass spectrometer coupled with an Agilent Technologies 1100 HPLC system. Buffer A (water with 0.2% formic acid) and buffer B (actonitrile with 0.2% formic acid) were used as mobile phases for the HPLC systems. Protein samples were loaded onto a Microtrap protein column (Michrom Bioresources Inc product 0001309) preequilibrated with 2% buffer B with a flow rate of 0.2 ml/min, and eluted with a gradient scheme: a linear increase to 70% buffer B in 8 min, then another linear increase to 98% buffer B in 2 min. The MS data were collected in positive ion mode by full MS scan from m/z 300 to 2000. The protein molecular weight was calculated by deconvoluting the mass spectrum with ProMass software. SDS–PAGE and Western blot All SDS–PAGE analysis was performed using NuPAGE 4–12% Bis–Tris Gels with the MES running buffer system (Invitrogen). Gels were either stained with Coomassie Blue or transblotted to PVDF membrane using the iBlot system (Invitrogen) for Western analysis. The Western blot was developed with Promega Western Blue method. Briefly, the PVDF membrane was incubated with the
Fig. 2. Expression strategies to improve BTK solubility and yield in E. coli cells. (A) Coexpression of chaperones affects soluble BTK level. Soluble fractions and total lysates are labeled as S and T, respectively. Samples loaded are equivalent to 20 ll culture. BTK alone (Lane 1) and coexpressed with chaperone groES-groEL-tig (Lane 2), Tig (Lane 3), groES-groEL (Lane 4), dnaK-dnaJ-grpE-groES-groEL (Lane 5) and dnaK-dnaJ-grpE (Lane 6). (B) Coexpression of YopH improves soluble BTK level by about onefold. Lane 1: soluble fraction from 2.5 ll BTK expressing sf21 cells; lane 2: soluble fraction from 20 ll E. coli culture with BTK expression alone; lane 3: soluble fraction from 20 ll E. coli culture with BTK and YopH coexpression. (C) Expression of BTK without codon optimization. Lane 1: total lysate from 20 ll wildt-type BTK expressing E. coli culture; lane 2: soluble fraction from 20 ll wild-type BTK expressing E. coli culture. (D) Expression of one codon optimized BTK. Lane 1: total lysate from 20 ll a codon optimized BTK expressing E. coli culture; lane 2: soluble fraction from 20 ll the codon optimized BTK expressing E. coli culture. (A and B) Western blotting results with anti-His6 antibody, (C and D) Coomassie Blue stained SDS–PAGE.
L. Wang et al. / Protein Expression and Purification 61 (2008) 204–211
primary antibody (all primary antibodies in this work were used in 1:1000 dilution in 5% Carnation nonfat milk in TBST buffer) after it was blotted with 5% Carnation nonfat milk in TBST buffer for 30 min. The membrane was washed three times with TBST (10 min/ wash). Then the membrane was probed with Promega AP-conju gated secondary antibody in TBST for 30 min. Then the membrane was developed with 5 ml Western Blue Stabilized Substrate for Alkaline Phosphatase (Promega) after being washed three times with TBST (10 min/wash). Results and discussion We constructed the idea to develop a method to obtain npPTKs while studying the activation mechanism of BTK in vitro. BTK is one of many PTKs shown to be activated by its tyrosine phosphoryla tion [20–22]. To understand the role of phosphorylation in human BTK kinase activation and its impact on small molecule inhibi tor discovery, we needed to develop expression and purification approaches which would allow us to obtain significant amounts of human full length BTK protein in the fully non-phosphorylated state (npBTK). An E. coli expression system was tried for its known lack of post-translational modification and literature precedence [23]. Unfortunately, we only obtained a low expression level in E. coli using the wild-type human BTK gene sequence, the major ity of which accumulated in inclusion bodies. Coexpressing with chaperone groES-groEL had increased soluble BTK in the lysate (Fig. 2A), but had no significant impact on the BTK yield after puri fication due to extra steps to eliminate co-purified chaperones. To address the problem of solubility, we attempted phosphatase coex
207
pression. Tyrosine phosphatase YopH has been shown to increase the solubility of protein tyrosine kinases Abl and Src in E. coli [15]. Using a similar method, we found that YOP coexpression only mar ginally increased BTK solubility with a low yield not sustainable for biochemical or structural biology studies (Fig. 2B). We were able to increase the BTK expression level more than tenfold with gene optimization (Fig. 2C and D); however the soluble BTK level remained low. Insect cell expression and post-expression phosphatase treat ment approaches were explored in parallel (Fig. 1B). As eukaryotic host cells, insect cells have been shown to be better host cells for human protein expression [7,24]. Baculovirus encoding hexa-His tine tagged full length human BTK was produced and infection of sf21 insect cells was performed using the Titerless Infected-cells Preservation and Scale-up (TIPS) method described previously [19]. Protein expression analysis suggested that human BTK protein expressed well in insect cells and the majority of the protein was in the soluble fraction. After metal affinity purification followed by ion-exchange chromatography (see Materials and methods for detail), we were able to obtain »4 mg BTK protein with more than 90% purity from 1 L sf21 cells at a density of 1.2 £ 106 cells/ml (Fig. 3A). When examined by Western analysis using pY antibody and mass spectroscopy, BTK obtained this way was found to be in a mix ture of highly heterogeneous phosphorylated states (Fig. 3B and D). We next tried to generate npBTK by treating purified BTK from the insect cells with a variety of phosphatases (k-PPase, YOP, TC PTP and CIP from New England BioLabs) followed by repurification. This approach turned out to be labor intensive and inefficient, with BTK often precipitating out of the solution.
Fig. 3. BTK from insect cells with or without coexpression with YopH. (A) Equal amount of purified BTK was analyzed on SDS–PAGE, stained with Coomassie Blue. Lane 1: 5 lg purified BTK without YopH coexpression; lane 2: 5 lg purified BTK with YopH coexpression. (B) Equal amount of purified BTK was analyzed on SDS–PAGE and Western-blot ted with pY antibody from Cell Signaling. Lane 1: 200 ng purified BTK without YopH coexpression; lane 2: 200 ng purified BTK with YopH coexpression. (C) Mass Spectrum analysis of purified BTK from insect cells with YopH coexpression. (D) Mass spectrum analysis of purified BTK from insect cells. The theoretical molecular weight of this construct is 81, 244 Da. BTK from insect cells without YopH coexpression is a mixture of protein at various phosphorylation stages with an average of two phosphorylation sites. “P” stands for one phosphorylation modification.
208
L. Wang et al. / Protein Expression and Purification 61 (2008) 204–211
Given the poor results of the post-expression treatment with phosphatases and the E. coli expression, we explored the coex pression of BTK with the tyrosine phosphatase YopH in insect cells for the following reasons: (1) insect cells are better hosts to express soluble kinases compared to E. coli cells [7]; (2) BTK is a protein tyrosine kinase and is known to be activated by its tyro sine phosphorylation (Y551 and Y223); (3) YopH is a phospha tase highly specific for phosphotyrosine [14]. To test this idea, we generated baculovirus encoding full length YopH with an Nterminal FLAG tag for coexpression. With our coexpression and purification procedure (see Materials and methods for detail), we routinely obtained »4 mg protein with >90% purity from 1 L sf21 cells (Fig. 3A), comparable to BTK yield without YopH coex pression. Data from Western-blot and mass spectrum analysis suggested that tyrosine residues in BTK from the insect cells with YopH coexpression were not phosphorylated (Fig. 3B and C). The observed molecular weight of the BTK protein from mass spectroscopy corresponded to the fully non-phosphorylated
full length BTK protein. The recombinant npBTK produced this way was shown to retain its biological function. It can undergo auto-phosphorylation reaction on multiple residues upon incu bation with ATP and Mg2+ in a time dependent fashion (Fig. 4A). Additionally the BTK upstream kinase Syk can further accelerate phosphorylation modification on the recombinant BTK (Fig. 4B). We have also examined the phosphorylation status on Y223 and Y551 residues, phosphorylation of which has been suggested to be essential for BTK to be fully active [20–22]. Western analy sis of the BTK auto-phosphorylation reaction using pY551 and pY223 specific antibodies suggested that both tyrosine resi dues can be phosphorylated in a time dependent manner and the phosphorylation on the Y551 residue appeared faster than the phosphorylation on Y223 residue (Fig. 4C and D). Using this approach and combined with site-directed mutagenesis, we have performed various kinetic and mechanism studies to explore the role of Y223 and Y551 phosphorylation in BTK activity regulation (results will be described in a separate manuscript).
Fig. 4. Activation analysis of purified npBTK. (A) Mass Spectrum analysis on the auto-phosphorylation of npBTK. npBTK was incubated with ATP/Mg2+ at room temperature, and samples were withdrawn at certain time points and analyzed with a mass spectrometer. (B) Mass Spectrum analysis on the phosphorylation of npBTK by upstream kinase Syk. npBTK was incubated with Syk and ATP/Mg2+ at room temperature, and samples were withdrawn at certain time points and analyzed with a mass spectrometer. (C and D) Western analysis on npBTK auto-phosphorylation with BTK pY551 and pY223 antibodies, respectively. Lanes 1, 2, 3, 4 and 5 are reactions at 0, 5, 10, 15 and 30 min, respectively. BTK (200 ng) was loaded in each lane. The details of the experiment are described in Materials and methods.
L. Wang et al. / Protein Expression and Purification 61 (2008) 204–211
PTKs have been shown to be the most difficult kinases to express in soluble form using E. coli [7]. Given our success with producing BTK using YopH insect cell coexpression method, we decided to test our insect cell YopH coexpression approach with other PTKs. We applied the method on additional human PTKs including cyto plasmic kinase JAK3 and receptor kinase EphA2. What is signifi cant is that this method worked equally well on these additional unrelated PTKs. The JAK3 kinase domain (JH1 domain, amino acid 813–1104) and the EphA2 intracellular catalytic domain (amino acid 560–976) were GST-tagged. Both proteins were purified with GST affinity chromatography followed by Heparin chromatography (see Materials and methods for detail), and analyzed with Western and mass spectroscopy. As shown in Fig. 5, expression of JAK3 alone resulted in heterogeneous phosphorylation on the protein (Fig. 5A and D), while coexpression with YopH phosphatase pro duced a completely de-phosphorylated JAK3 (Fig. 5B and C). The molecular weight of the major peak agreed well with the calcu lated non-phosphorylated JAK3 with N-Met removal, the molecu lar weight of the minor peak was equal to the theoretical molecu lar weight of acetylated JAK3. Similar results were observed with EphA2 recombinant protein (Fig. 6). There were two more minor peaks (MW 74, 822 and 74, 906) other than the major non-phos phorylated EphA2 peak (MW 74, 739) in the mass spectroscopy (Fig. 6C) of YopH coexpression product. Further studies are needed to understand whether these are Ser/Thr phosphorylation states of the EphA2 protein when expressed in insect cells. Compared with the enzymatic post-expression treatment (Fig. 1B), the baculovirus mediated coexpression (Fig. 1C) has several
209
advantages besides cost. First, it greatly simplifies the overall pro cess which is evident in Fig. 1, and does not require enzymatic treatment and re-purification. Second, post-expression enzymatic treatment is an optimization process like any assay development. It needs a lot of work to identify the optimal condition. Third, post-expression enzymatic treatment requires prolong incubation of purified protein on ice or room temperature. The protein stabil ity will limit this approach. Fourth, sometimes there is no easy translation from small scale test to scale-up for post-expression treatment. This actually was the bottle neck when we treated BTK with phosphatases. With the baculovirus mediated coexpression, certain phosphatases might also affect the expression and folding ability of insect cells therefore causing some undesired results. Fortunately it is not the case for YopH phosphatase. With all PTKs tested, the yield for npPTKs stays the same if not better. An alternative option to coexpression is to use virus generated from a single plasmid carrying both PTKs and YopH using duel vec tors (pFastBac DUAL etc). In principle, it should produce npPTKs. The downsides are (1) it only produces npPTKs; (2) for every PTK, one dual expression construct needs to be made; (3) one will lose the flexibility to adjust kinase-phosphatase ratio necessary to achieve optimal expression condition. On the other hand, our strategy will provide the flexibility to adjust kinase:phosphatase ratio to achieve optimal npPTK yield, to produce PTK or npPTK with only one YopH construct. Therefore it provides flexibility as well as cost efficiency. In summary, we report a method to obtain high yield expression of npPTKs from insect cells with YopH coexpression. Coexpression of kinase and phosphatase in mammalian cells has long been used
Fig. 5. JAK3 from insect cells with or without coexpression with YopH. (A) Equal amount of purified JAK3 was analyzed on SDS–PAGE, stained with Coomassie Blue. Lane 1: 5 lg purified JAK3 without YopH coexpression; lane 2: 5 lg purified JAK3 with YopH coexpression. (B) Equal amount of purified JAK3 was analyzed on SDS–PAGE and Western-blotted with pY antibody from Cell Signaling. Lane 1: 200 ng purified JAK3 without YopH coexpression; lane 2: 200 ng purified JAK3 with YopH coexpression. (C) Mass Spectrum analysis of purified JAK3 with YopH coexpression. (D) Mass Spectrum analysis of purified JAK3 from insect cells without YopH coexpression. The theoretical molecular weight of this construct is 59, 642 Da or 59, 511 Da (N-terminal Met removal).
210
L. Wang et al. / Protein Expression and Purification 61 (2008) 204–211
Fig. 6. EphA2 from insect cells with or without coexpression with YopH. (A) Equal amount of purified EphA2 was analyzed on SDS–PAGE, stained with Coomassie. Lane 1: 5 lg purified EphA2 without YopH coexpression; lane 2: 5 lg purified EphA2 with YopH coexpression. (B) Equal amount of purified EphA2 was analyzed on SDS–PAGE and Western-blotted with pY antibody from Cell Signaling. Lane 1: 200 ng purified EphA2 without YopH coexpression; lane 2: 200 ng purified EphA2 with YopH coexpression. (C) Mass Spectrum analysis of purified EphA2 from insect cells with YopH coexpression. (D) Mass Spectrum analysis of purified EphA2 from insect cells without YopH coex pression. The theoretical molecular weight of this construct is 74, 906 Da. The molecular weight of unphosphorylated peaks (74, 739 Da in C and 74, 742 Da in D) »165 Da less than the theoretical molecular weight may be due to other modifications and have no simple explanation.
to study the kinase activation mechanism and elucidate its biolog ical function. It has also been used in E. coli expression systems to generate non-phosphorylated kinases or improve kinase yield [8,15,25,26]. To our knowledge, this is the first report to employ kinase/phosphatase coexpression strategy from insect cells to produce npPTKs. Taking advantage of the insect cell’s eukaryotic expression system, which has an excellent track record for gener ating human kinases, we were able to produce functional and sol uble npPTKs through coexpression with the tyrosine phosphatase YopH. This method was tested against cytoplasmic tyrosine kinases as well as receptor tyrosine kinase, and achieved equally positive results. These results suggest that the insect cell YopH coexpression method can be generally applied to other tyrosine kinases. It greatly simplifies the process to generate npPTKs, and will contribute to bet ter understanding the mechanism of the tyrosine kinase activation process and the discovery of kinase inhibitors with novel inhibition mechanisms. It is reasonable to believe that a similar process can be applied to generate non-phosphorylated Ser/Thr kinases in insect cells by coexpressing them with a Ser/Thr phosphatase. Acknowledgments We thank Alexis C. Meng for her help in preparing of the manu script, Robert Stanton for his helpful inputs. References [1] M.E. Noble, J.A. Endicott, L.N. Johnson, Protein kinase inhibitors: insights into drug design from structure, Science 303 (2004) 1800–1805.
[2] J.S. Sebolt-Leopold, J.M. English, Mechanisms of drug inhibition of signaling molecules, Nature 441 (2006) 457–462. [3] S.R. Hubbard, L. Wei, L. Ellis, W.A. Hendrickson, Crystal structure of the tyrosine kinase domain of the human insulin receptor, Nature 372 (1994) 746–754. [4] S.R. Hubbard, Crystal structure of the activated insulin receptor tyrosine kinase in complex with peptide substrate and ATP analog, EMBO J. 16 (1997) 5572–5581. [5] S.R. Hubbard, Protein tyrosine kinases: autoregulation and small-molecule inhibition, Curr. Opin. Struct. Biol. 12 (2002) 735–741. [6] T. Schindler, W. Bornmann, P. Pellicena, W.T. Miller, B. Clarkson, J. Kuriyan, Structural mechanism for STI-571 inhibition of Abelson tyrosine kinase, Sci ence 289 (2000) 1938–1942. [7] S.P. Chambers, D.A. Austen, J.R. Fulghum, W.M. Kim, High-throughput screen ing for soluble recombinant expressed kinases in Escherichia coli and insect cells, Protein Expr. Purif. 36 (2004) 40–47. [8] R.A. Elling, B.T. Tangonan, D.M. Penny, J.T. Smith, D.E. Vincent, S.K. Hansen, T. O’Brien, M.J. Romanowski, Mouse Aurora A: expression in Escherichia coli and purification, Protein Expr. Purif. 54 (2007) 139–146. [9] P. Du, P. Loulakis, C. Luo, A. Mistry, S.P. Simons, P.K. LeMotte, F. Rajamohan, K. Rafidi, K.G. Coleman, K.F. Geoghegan, Z. Xie, Phosphorylation of serine res idues in histidine-tag sequences attached to recombinant protein kinases: a cause of heterogeneity in mass and complications in function, Protein Expr. Purif. 44 (2005) 121–129. [10] C.K. Smith, D. Carr, T.W. Mayhood, W. Jin, K. Gray, T.W. Windsor, Expression and purification of phosphorylated and non-phosphorylated human MEK1, Protein Expr. Purif. 52 (2007) 446–456. [11] P.M. Chan, S. Ilangumaran, J. La Rose, A. Chakrabartty, R. Rottapel, Autoinhi bition of the Kit receptor tyrosine kinase by the cytosolic juxtamembrane region, Mol. Cell. Biol. 23 (2003) 3067–3078. [12] J.H. Till, M. Becerra, A. Watty, Y. Lu, Y. Ma, T.A. Neubert, S.J. Burden, S.R. Hub bard, Crystal structure of the MuSK tyrosine kinase insights into receptor auto regulation, Structure 10 (2002) 1187–1196. [13] K.L. Guan, J.E. Dixon, Protein tyrosine phosphatase activity of an essential viru lence determinant in Yersinia, Science 249 (1990) 553–556. [14] Z.Y. Zhang, J.C. Clemens, H.L. Schubert, J.A. Stuckey, M.W. Fischer, D.M. Hume, M.A. Saper, J.E. Dixon, Expression, purification, and physicochemical character ization of a recombinant Yersinia protein tyrosine phosphatase, J. Biol. Chem. 267 (1992) 23759–23766.
L. Wang et al. / Protein Expression and Purification 61 (2008) 204–211
[15] M.A. Seeliger, M. Young, M.N. Henderson, P. Pellicena, D.S. King, A.M. Falick, J. Kuriyan, High yield bacterial expression of active c-Abl and c-Src tyrosine kinases, Protein Sci. 14 (2005) 3135–3139. [16] C.I. Smith, B. Baskin, P. Humire-Greiff, J.N. Zhou, P.G. Olsson, H.S. Maniar, P. Kjellen, J.D. Lambris, B. Christensson, L. Hammarstrom, Expression of Bruton’s agammaglobulinemia tyrosine kinase gene, BTK, is selectively down-regulated in T lymphocytes and plasma cells, J. Immuol. 152 (1994) 557–565. [17] S.G. Rane, E.P. Reddy, JAK3: a novel JAK kinase associated with terminal differ entiation of hematopoietic cells, Oncogene 9 (1994) 2415–2423. [18] R.A. Lindberg, T. Hunter, cDNA cloning and characterization of eck, an epi thelial cell receptor protein-tyrosine kinase in the eph/elk family of protein kinase, Mol. Cell. Biol. 10 (1990) 6316–6324. [19] D.J. Wasilko, E. Lee, TIPS: titerless infected-cells preservation and scale-up, Bio Processing J. 5 (3) (2006) 29–32. [20] D.E. Afar, H. Park, B.W. Howell, D.J. Rawlings, J. Cooper, O.N. Witte, Regulation of Btk by Src family tyrosine kinases, Mol. Cell. Biol. 16 (1996) 3465–3471.
211
[21] H. Park, M.I. Wahl, D.E. Afar, C.W. Turck, D.J. Rawlings, C. Tam, A.M. Scharen berg, J.P. Kinet, O.N. Witt, Regulation of BTK function by a major phosphoryla tion site within the SH3 domain, Immunity 4 (1996) 515–525. [22] D.J. Rawlings, A.M. Scharenberg, H. Park, M.I. Wahl, S. Lin, R.M. Kato, A.C. Fluck iger, O.N. Witte, J.P. Kinet, Activation of BTK by a phosphorylation mechanism initiated by SRC family kinases, Science 291 (1996) 822–825. [23] K. Bence, W. Ma, T. Kozasa, X.Y. Huang, Direct stimulation of Bruton’s tyrosine kinase by Gq-protein a-subunit, Nature 389 (1997) 296–299. [24] J.M. Vlak, R.J. Keus, Baculovirus expression vector system for production of viral vaccines, Adv. Biotechnol. Process. 14 (1990) 91–128. [25] T. Matsui, K. Tanihara, T. Date, Expression of unphosphorylated form of human double-stranded RNA-activated protein kinase in Escherichia coli, Biochem. Bio phys. Res. Commun. 284 (2001) 798–807. [26] W. Wang, A. Marimuthu, J. Tsai, A. Kumar, H.I. Krupka, C. Zhang, B. Powell, Y. Suzuki, H. Nguyen, M. Tabrizizad, C. Luu, B.L. West, Structural characterization of autoinhibited c-Met kinase produced by coexpression in bacteria with phos phatase, Proc. Natl. Acad. Sci. USA 103 (2006) 3563–3568.