baculovirus expression system: an alternative for ...

МОЛЕКУЛЯРНАЯ БИОЛОГИЯ, 2010, том 44, № 3, с. 535–540

ДРУГИЕ ПРОБЛЕМЫ UDK 577.21

BACULOVIRUS EXPRESSION SYSTEM: AN ALTERNATIVE FOR PRODUCING CATALYTICALLY ACTIVE HUMAN PTP1B © 2010 Sindhuja Sundaram1, Prabhakar Tiwari1, Shalini Saini1, Rajiv Kant1, Joseph Alex Davis2, Sudhir Sahdev1*, Kulvinder Singh Saini1 1

Department of Molecular Technology, New Drug Discovery Research, Ranbaxy Research Laboratories, Gurgaon122001, India Department of Pharmacology, New Drug Discovery Research, Ranbaxy Research Laboratories, Gurgaon122001, India

2

Received 19.08.2009 Accepted for publication 06.10.2009

Protein tyrosine phosphatases (PTPs) play multiple roles in many physiological processes. Overexpression of the PTPs has been shown to be associated with cellular toxicity, which may also lead to the deletion of the respective gene from stable cell clones. We also observed that PTP1B overexpression in CHO and HEK293 stable cell clones led to cytotoxicity and low revival rates during clone generation and maintenance. To address these issues, bacmid transpo sition technology was utilized to generate recombinant PTP1B baculovirus, and Spodoptera frugiperda (Sf9 and Sf21) insect cell lines were infected with the virus. The data obtained on expression and activity of the PTP1B high lights clear advantage of the recombinant baculovirusinsect cell expression system over the mammalian cell line tech nique due to increase in enzyme activity, strongly inhibited by phosphatase specific inhibitor RK682. Possible applica tion of the expression system for producing active enzymes in bulk quantity for a new drug discovery is also discussed. Key words: bacmid, pNPP assay, PTPactivity, RK682, transposition.

INTRODUCTION Protein tyrosine phosphatases (PTPs) are a diverse family of proteins constituting mainly of the transmem brane enzymes like leukocyte antigenrelated (LAR) PTP receptor, PTP receptor typeC (CD45) and the cytoplas mic enzymes like PTP1B, PTPSYP (SHP2), Tcell PTP (TCPTP), and etc. [1]. In many physiological pro cesses including cell growth, differentiation, mitosis, on cogenic transformation and down regulation of insulin signaling, these enzymes dephosphorylate various signal ing molecules [2–4]. PTP1B, a 50kDa protein of ty rosine phosphatase family, is usually localized in the en doplasmic reticulum [5] and was expressed in bacterial, mammalian and baculovirus systems. Over the years, PTP1B is a potent drug target for typeII diabetes, since it acts as a negative regulator of insulin signaling. From crystallographic studies, two important amino acids, Cys215 and Asp181, which are responsible for substrate binding and catalytic activity, were identified in the PTP signature motif and WPD (TrpProAsp) loop, respec tively. The Cys residue in the PTP1B catalytic domain gets oxidized even upon mild air exposure during process ing [6]. Since oxidation is known to be limiting factor in terms of activity, considerable precautions should be taken in course of PTP1B enzymatic activity assays. Abbreviations: h.p.i. – hours post infection; pNPP – pnitrophenyl phosphate; PTP – protein tyrosine phosphatase; RK682 – 3hexa decanoyl5hydroxymethyltetronic acid; RU – relative units. * email: [email protected]

It was shown that PTP1B expression in mammalian cells NIH3T3 failed to yield an enzyme with more than 2⎯3fold activity [7, 8]. In our laboratory, we generated HEK293 stable cells expressing high levels of the PTP1B displaying 6fold higher activity for the initial few passages than that for the control. However, increasing the passage number during subculturing led to development of signif icant cytotoxicity. In the case of CHOPTP1B stable cell clone culture, a low expression level and enzyme activity was another key concern. The problem mentioned per suaded us to look for alternative expression systems. Use of the baculovirusinsect cell system for PTP1B expres sion allowed us to achieve significantly higher expression level and yield of the biologically active enzyme at the negligible cellular cytotoxicity induced. EXPERIMENTAL Restriction and modifying enzymes were purchased from New England Biolabs (USA). The CHO and HEK293 cell lines were from ATCC (USA). Sf9, Sf21 cells and the BactoBac vector expression system were obtained from Invitrogen (USA). Both antiHis antibody and the antimouse antibodies were from Sigma (USA). The substrate pNPP and the phosphatase specific inhibi tor RK682 were from Fluka (USA) and Biomol (USA), respectively. PTP1B cloning and expression in mammalian cells. Proof reading Phusion DNA polymerase (NEB, USA)

535

536

SINDHUJA SUNDARAM et al. 1

2

3

4

5

6

7

8 60 kDa

Fig. 1. PTP1B expression in the mammalian stable cell clones. Western blot assay of the recombinant PTP1B ex pression with use of antiHis antibodies: 1 – protein mark er; 2 – HEK293control; 3 – HEKPTP1B clone 73; 4 – CHOcontrol; 5–8 – CHOPTP1B clones 21, 15, 11 and 6, respectively.

was used for PCR amplification of the fulllength PTP1B gene (GeneID: 5770) from the Image clone (pCMV.SPORT6/PTP1B, Origene technologies, USA). The primers used for PCR amplification (BamHI and AgeI sites deliberately inserted for cloning are under lined) were the following: PTP1B forward primer: 5'CGTGGATCCTGGC CCGTCATGGAGATG 3' and PTP1B reverse primer: 5'GCTACCGGTTGTGT TGCTGTTGAACAGGAACCTG3'. PCR amplification was performed for 30 cycles with DNA denaturation at 94°C for 30 s, annealing at 55°C for 30 s, and polymerization at 72°C for 1 min with a 2min preheating at 94°C followed by the final extension at 72°C for 5 min. The resulting PCR product was gel purified, di gested with the BamHI and AgeI restriction enzymes and cloned in a pGene/V5HisA vector (Invitrogen, USA) to get an inframe Cterminal histidine tagged PTP1B fu sion construct. The PTP1B positive clones were picked out after antibiotic selection, analyzed by restriction di gestion, and the sequence of the inserts was verified. CHO and HEK293 cell lines harboring the pSwitch vector were propagated in Ham’s F12K and MEM me dia, respectively, and supplemented with 10% heat inacti vated fetal bovine serum. These cells were subsequently transfected with 1 μg of the linearized pGene/PTP1B construct using the Lipofectamine 2000 (Invitrogen, USA) as described by the manufacturer. Single cell clones were selected and expanded using increasing concentra tions of Zeocin and Hygromycin up to 1 mg and 0.2 mg, respectively. 25 clones were analysed for PTP1B expres sion by Western blotting with antiHis antibodies and for PTPactivity by the pNPP hydrolysis assay. PTP1B cloning and expression in baculovirus system. Recombinant baculoviral DNA harboring human PTP 1B cDNA was cloned in the pFastBac1 vector of the bac ulovirus expression system. The pGenePTP1B clone was digested with the BamHI and PmeI restriction en zymes in order to release the PTP1B histidine tag cas sette and then subcloned into the pFastBac1. The DH10Bac E. coli cells (Invitrogen, USA) were trans formed with the recombinant pFastBac1PTP1B plas mid. The recombinant bacmid DNA formation is facili tated by transposonmediated exchange of the expression cassette, which disrupts the lacZα gene of the parent bac

mid leading to the formation of whitecolored colonies. The recombinant PTP1B clones were checked for the presence of the gene specific insert by bluewhite screen ing, and the pure white clones were analyzed by colony PCR using the M13 forward and reverse primers. The PCR positive DH10BacPTP1B colonies were cultured overnight to isolate the PTP1B bacmid DNA by the al kaline lysis method. The Sf9 and Sf21 insect cells were cultured in SF900 serumfree media (Invitrogen, USA) at 27°C in a BOD incubator. The cells (2.5 × 106) were transfected with PTP1B and wild type bacmid DNA (2 μg of each) in a 6 well plate using Cellfectin (7 μl per well; Invitrogen, USA). Transfected cells were monitored for infection by microscopy. Wild type baculoviruses as well as the PTP 1B recombinants, harvested at 96 h post infection (h.p.i.), were used for further amplification according to the man ufacturer’s instruction. The PTP1B expression at various time points after the infection was analyzed by Western blotting. PTP1B enzyme activity for pNPP substrate. PTP1B expressing cells were centrifuged and the pellets were lysed for 15 min at 37°C using the cell lysis buffer (Sigma, USA) and then centrifuged at 4°C for 15 min as per the suppli er’s instruction. A PTP1Bcontaining supernatant was carefully aliquoted into nitrogen preflushed tubes to avoid oxidation and then used for the enzymatic activity assay. For measuring the PTPactivity the pNPP was used as a substrate. The enzyme was preincubated in the assay buffer (50 mM NaOAc, 1mM EDTA, 10 mM DTT and 0.05% Igepal; pH 5.5) at 30°C for 10 min, followed by the addition of the substrate and incubation the mixture at 30°C for 20 min. The reaction was stopped by adding 1N solution of NaOH that resulted in development of yellow color due to an appearance of the deprotonated phenyl group released by the phosphatase from pNPP molecules. Finally, absorbance was measured at 405 nm and the spe cific activity of PTP1B enzyme was calculated as pmoles/min/mg. RESULTS PTP1B activity in mammalian stable cell clone lysates Lysates of the HEK293PTP–1B stable subclones 73.41 and 73.31(expression of the recombinant PTP–1B was confirmed by Western blot analysis using antiHis an tibodies, Fig. 1) displaying 3.5 and 6.0fold higher activ ity as compared to the nontransfected control, respec tively, were selected for the further analysis. Moreover, the PTPactivity of these clones was inhibited (up to 79%) by a phosphatasespecific inhibitor RK682 (Fig. 2). Howev er, the HEK293PTP1B stable cell clones could not be revived from their respective frozen stocks. Basal phosphatase activity level in the CHO cell lysates was higher (4083 pmoles/min/mg) than that in the МОЛЕКУЛЯРНАЯ БИОЛОГИЯ

том 44

№3

2010

BACULOVIRUS EXPRESSION SYSTEM: AN ALTERNATIVE FOR PRODUCING

PTPactivity (RU)

а

б

7 6 5 4 3 2 1 0 Clone 73.31

537

Clone number

Inhibition, %

73.31

79

73.41

75

HEK293 (Control)

13

Clone 73.41 HEK293 control

Fig. 2. Enzymatic activity of the PTP1B expressed in the HEK293 stable cell clones. a – PTP1B activity of the HEK293 stable cell clone lysates related to that for the nontransfected control (n = 1) presented in relative units (RU); b – inhibition of PTP 1B activity of the HEK293 S stable cell clone lysates by the RK682.

PTP–activity (RU)

2.5 2.0 1.5 1.0 0.5 0 Clone 6

Clone 11

Clone 15

Clone 21

CHO control

Fig. 3. Enzymatic activity of the PTP1B expressed in CHO stable cell clones. A PTP1B activity of the CHO stable cell clone lysates from initial passages (grey bars) and corresponding repeated subcultured clones (white bars) related to the nontransfected control (n = 1) is presented in relative units (RU).

HEK293 cells (883 pmoles/min/mg). Out of 25 stable clones of PTP1B transfected CHO cells only the clone 21 lysate displayed 2fold higher enzymatic activity than the nontransfected control one; the phosphatase activity was inhibited up to 53% by RK682. However, after follow ing repeated passages of the selected CHOPTP1B clones the enzymatic activity of the lysates was 2–5 times lower than that for the initial passages (Fig. 3) and not in hibited by RK682.

tive/background controls. A maximal level of recombi nant PTP1B in insect cells infected with the recombi nant baculovirus was observed at 80 h.p.i. for the Sf21 cells and at 48 h.p.i. for the Sf9 cells (Fig. 4). It should be

a 2

1

3 60 kDa

Time course of the PTP1B expression in the recombinant baculovirus infected insect cell system The recombinant human PTP1B baculovirus gener ated with the pFastBac1 shuttle vector cloning was used for the subsequent expression studies. PTP1B expres sion was analyzed at different time intervals (h.p.i.) by Western blot using antiHis antibodies. The recombinant PTP1BHistag protein migration on a 10% SDS PAGE corresponded to the expected molecular mass (60 kDa). Lysates of insect cells Sf9 and Sf21 infected with the wild type baculovirus (wt) were used as nega МОЛЕКУЛЯРНАЯ БИОЛОГИЯ

том 44

№3

2010

b 1

2

3

4 60 kDa

Fig. 4. Western blot assay of recombinant PTP1B expres sion in the baculovirus infected insect cells with use of an tiHis antibodies. The data present PTP1B level at vari able time points after the PTP1B baculovirus infection (h.p.i.) of insect cells: a – control, Sf21wt at 24 h.p.i. (1), Sf21PTP1B at 96 h.p.i. (2), Sf21PTP1B at 80 h.p.i. (3); b – Sf9PTP1B at 72 h.p.i. (1), Sf9PTP1B at 48 h.p.i. (2), Sf9PTP1B at 24 h.p.i. (3), control Sf9wt at 48 h.p.i. (4).

538

SINDHUJA SUNDARAM et al. 12 Sf21

Sf9

PTPactivity (RU)

10 8 6 4 2 0 24 h

40 h

48 h

65 h Control 18 h

24 h

48 h Control

Hours post infection

Fig. 5. The time course of PTP1B enzymatic activity expressed in the recombinant baculovirusinsect cell lysates. A basal PTP 1B activity level of the wild type baculovirus infected Sf9 and Sf21 insect cells was measured at 48 h.p.i. (control); a PTP1B ac tivity of the recombinant baculovirus infected Sf9 and Sf21 cells at the time indicated as the h.p.i. is presented in relative units to the control (n = 1). The data of 2 experiments are shown by grey and white bars.

noted that basal PTPlevel in the Sf9 cells was higher than that in the Sf21 cells. PTP1B activity of insect cell lysates The Sf9 and Sf21 insect cells were independently in fected with the wild type and PTP1Bcontaining viruses at a multiplicity of infection (m.o.i.) of 1. Cell lysates were prepared from the infected cells gathered at various time points after infection. A phosphatase activity of the PTP 1B infected Sf21 cells at 48 h.p.i. was 4–5 times higher than that of the wt virus infected control; the difference further increasing up to ratio of 8–10 at 65 h.p.i. (Fig. 5). After that the PTP1B activity declined, possibly because of development of a cytotoxicity induced at the late phase of the viral infection (data not shown). The time course of PTP1B activity in the Sf21 infect ed cells was almost linear, whereas in the Sf9 cells the de pendence was mild that may be associated with the high basal phosphatase activity level in this cell line. The PTP1B harboring baculovirus was used many times during 6 months to check for any changes in a viral titer and expression level of the enzyme. Consistent PTP 1B activity and RK682mediated inhibition in the PTP 1Bcontainng virus stocks during their longterm storage was also observed (data not shown). DISCUSSION Protein tyrosine phosphorylation is a reversible dy namic process where the level of phosphorylation is the critical ratelimiting step for controlling cell growth and differentiation. PTP1B has been known to negatively regulate cell proliferation by specifically interfering with celladhesionmediated signaling pathways via proline

dependent interactions with one or more critical compo nents of the adhesiondependent signaling apparatus [3]. Certain enzymes of the PTP class are potential therapeu tic targets in a variety of diseases such as diabetes, inflam mation and cancer [9]. PTP1B is an important target for the treatment of diabetes, since PTP1B null mice were shown to develop a resistance to dietinduced diabetes and obesity [4]. Members of the phosphoprotein phosphatase (PPP) family of serine/threonine phosphatases, including pro tein phosphatases PP1, PP2A and PP2B, share invariant active site residues that are critical for the catalytic func tion. Various PTP member proteins, GST tagged psiPTEN [10], human receptor tyrosine phosphatase gamma gene (PTPRG) and other phosphatases were overexpressed in baculovirus insect cell systems and structurally characterized after purification [11]. Biologi cally active recombinant human PP2Acα, PP2Acβ and wild type systems as well as an activesite mutant of PP2Ac were also well characterized in the recombinant baculovirusHigh Five insect cells [12, 13]. PTPs invariably carry a Cys, which needs to be in a re duced state to remain catalytically active. Since exposure to air leads to oxidation of this highly sensitive Cys resi due, a PTP activity assay is normally carried out in the presence of potent reducing agents. However, this ap proach is not used for the activity assay of endogenous PTPs, which are directly isolated from the cellular envi ronment. Keeping these observations in mind, utilization of an anaerobic chamber during the PTP1B lysate prep aration was shown to prevent the air mediated oxidation of total PTPs [14, 15]. The human PTP1B was also overexpressed in mam malian cell lines including NIH3T3 (8) and baby ham МОЛЕКУЛЯРНАЯ БИОЛОГИЯ

том 44

№3

2010

BACULOVIRUS EXPRESSION SYSTEM: AN ALTERNATIVE FOR PRODUCING

ster kidney (BHK) cells [16]. The New Chemical Entity (NCE) screening assays require regular production of the target enzyme, therefore continuous propagation of the PTP1B expressing stable cell clones becomes prerequi site. However, after a few passages both maintenance of the mammalian PTP1B stable cell clones and secondary clonal selection process became difficult. The PTP1B enzyme activity in the cells of the late passages (e.g. pas sage 5 onwards) was significantly reduced. Apart from this, repeated attempts at reviving these active PTP1B cell clones from the cryopreserved stocks proved to be fu tile. During revival of the clones cell death was markedly increased that may be a result of recombinant PTP1B overexpression. On the basis of the known literature data and the result obtained we concluded that PTP1B over expression may result in an upsurge of the phosphatase ac tivity in cellular milieu leading to cytotoxicity and subse quent failure of its proper and consistent expression [6]. The baculovirusinsect cell expression system provides an efficient way to circumvent these problems. PTP1B production can be consistently maintained through fresh infections using the recombinant baculovirus stocks, which are found to be stable even during corresponding longterm storage. This strategy can very well counter the shortcomings associated with the basal expression levels of the toxic enzymes, observed during mammalian cell clone work. Apart from this, since the recombinant baculovirus infection is carried out from original stocks, the specific activity of the expressed enzyme also remains consistent. There are earlier reports, which have specifically dealt with the expression and functional characterization of PTP1B in the baculovirusSf9 insect cell system. It was observed that for the recombinant baculovirus infected Sf9 cells expressed fulllength PTP1B was processed and also localized to the predicted subcellular compartments [17]. However, as this work was particularly focused on analyzing the PTP1B activity at a very early h.p.i. phase, the role of the specific insecthost as well as baculovirus specific phosphatase background generated during the late phase of the infection cycle was not taken into consid eration. We compared the Sf9 and Sf21 cells and found that the later provide us with much better options for producing biologically active PTP1B due to the significantly lower expression levels of the intrinsic hostcell and baculovirus specific phosphatases. This is an important observation, as it provides us with an advantage during screening of the PTP1B specific inhibitors by offering broader activity windows between the wtinfected controls and the PTP 1B expressing cells, and thereby considerably reducing the basal activity levels. Our strategy of expressing com plex proteins/enzymes through baculovirusinsect cell systems is quite valuable, since it saves time and is relative ly inexpensive. In addition, this host is quite effective in counteracting cytotoxicity problems specific to protein МОЛЕКУЛЯРНАЯ БИОЛОГИЯ

том 44

№3

2010

539

expression and also provides the scientists searching new drugs with a significantly higher activity window as com pared to the mammalian expression systems for screening of low molecular mass inhibitors. These critical findings point toward the advantages of using the baculovirusin sect expression system over the existing mammalian ex pression system for producing therapeutically important target proteins. ACNOWLEDGEMENTS This work was funded by the Ranbaxy Research Labo ratories, India. We thankfully acknowledge our colleagues within the New Drug Discovery Research Group, the Department of Molecular Technology and Department of Pharmacology including Dr. G.V. Rayasam for their in put and discussions. COMPETING INTERESTS STATEMENT The authors declare no competing interests. REFERENCES 1. Dadke S., Kusari J., Chernoff J. 2000. Downregulation of insulin signaling by proteintyrosine phosphatase 1B is me diated by an Nterminal binding region. J. Biol. Chem. 275, 23642–23647. 2. Sorenson C.M., Sheibani N. 2002. Altered regulation of SHP2 and PTP 1B tyrosine phosphatases in cystic kidneys from bcl2 / mice. Am. J. Physiol. Renal Physiol. 282, F442–450. 3. Liu F., Sells M.A., Chernoff J. 1998. Protein tyrosine phos phatase 1B negatively regulates integrin signaling. Curr. Bi ol. 8, 173–176. 4. Dube N., Bourdeau A., Heinonen K.M., Cheng A., Loy A.L., Tremblay M.L. 2005. Genetic ablation of pro tein tyrosine phosphatase 1B accelerates lymphomagenesis of p53null mice through the regulation of Bcell develop ment. Cancer Res. 65, 10088–10095. 5. Frangioni J.V., Beahm P.H., Shifrin V., Jost C.A., Neel B.G. 1992. The nontransmembrane tyrosine phosphatase PTP1B localizes to the endoplasmic retic ulum via its 35 amino acid Cterminal sequence. Cell. 68, 545–560. 6. Yang J., Liang X., Niu T., Meng W., Zhao Z., Zhou G.W. 1998. Crystal structure of the catalytic domain of pro teintyrosine phosphatase SHP1. J. Biol. Chem. 273, 28199–28207. 7. Yang J., Liu L., He D., Song X., Liang X., Zhao Z.J., Zhou G.W. 2003. Crystal structure of human protein tyrosine phosphatase SHP1. J. Biol. Chem. 278, 6516– 6520. 8. Moeslein F.M., Myers M.P., Landreth G.E. 1999. The CLK family kinases, CLK1 and CLK2, phosphorylate and activate the tyrosine phosphatase, PTP1B. J. Biol. Chem. 274, 26697–26704. 9. Huang P., Ramphal J., Wei J., Liang C., Jallal B., McMa hon G., Tang C. 2003. Structurebased design and discov ery of novel inhibitors of protein tyrosine phosphatases. Bioorg. Med. Chem. 11, 1835–1849.

540

SINDHUJA SUNDARAM et al.

10. Fujii G.H., Morimoto A.M., Berson A.E., Bolen J.B. 1999. Transcriptional analysis of the PTEN/MMAC1 pseudogene, psiPTEN. Oncogene. 18, 1765–1769. 11. Sorio C., Mendrola J., Lou Z., LaForgia S., Croce C.M., Huebner K.1995. Characterization of the receptor protein tyrosine phosphatase gene product PTP gamma: binding and activation by triphosphorylated nucleosides. Cancer Res. 55, 4855–4864. 12. Ikehara T., Shinjo F., Ikehara S., Imamura S., Yasumoto T. 2006. Baculovirus expression, purification, and character ization of human protein phosphatase 2A catalytic subunits alpha and beta. Protein Expr. Purif. 45, 150–156. 13. Myles T., Schmidt K., Evans D.R., Cron P., Hem mings B.A. 2001. Activesite mutations impairing the cata lytic function of the catalytic subunit of human protein

14. 15. 16.

17.

phosphatase 2A permit baculovirusmediated overexpres sion in insect cells. Biochem. J. 357, 225–232. Zhang Z.Y. 1998. Proteintyrosine phosphatases: biologi cal function, structural characteristics, and mechanism of catalysis. Crit. Rev. Biochem. Mol. Biol. 33, 1–52. Denu J.M., Dixon J.E. 1998. Protein tyrosine phosphatas es: mechanisms of catalysis and regulation. Curr. Opin. Chem. Biol. 2, 633–641. Cool D.E., Tonks N.K., Charbonneau H., Fischer E.H., Krebs E.G. 1990. Expression of a human Tcell protein tyrosinephosphatase in baby hamster kidney cells. Proc. Natl. Acad. Sci. USA. 87, 7280–7284. Cromlish W.A., Payette P., Kennedy B.P. 1999. Devel opment and validation of an intact cell assay for protein tyrosine phosphatases using recombinant baculovirus es. Biochem. Pharmacol. 58, 1539–1546.

МОЛЕКУЛЯРНАЯ БИОЛОГИЯ

том 44

№3

2010