In vitro selection of scFv and its production: an application of mRNA display and wheat embryo cell-free and E. coli cell production system

Please download to get full document.

View again

of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Document Description
In vitro selection of scFv and its production: an application of mRNA display and wheat embryo cell-free and E. coli cell production system
Document Share
Document Tags
Document Transcript
  APPLIED GENETICS AND MOLECULAR BIOTECHNOLOGY In vitro selection of scFv and its production: an applicationof mRNA display and wheat embryo cell-free and  E. coli   cellproduction system Tatsuro Shibui  &  Teruaki Kobayashi  & Keiichiro Kanatani  &  Hirohisa Koga  &  Satoru Misawa  & Tetsu Isomura  &  Tooru Sasaki Received: 6 February 2009 /Revised: 8 April 2009 /Accepted: 12 April 2009 /Published online: 7 May 2009 # Springer-Verlag 2009 Abstract  Synthetic DNA libraries encoding human anti- body V L  and V H  fragments were designed, constructed, andenriched using mRNA display. The enriched libraries werethen combined to construct a scFv library for mRNAdisplay. Sequencing revealed that 46% of the library codedfor full-length scFvs. Considering the number of moleculesused in mRNA display, the size of the library displayed wascalculated to be >10 10 . To verify this, we tried to isolate ascFv against human RANK. A scFv was successfullyisolated in the sixth round of panning and was synthesizedin wheat embryo cell-free (WE) and  Escherichia coli  cellsystems. In the WE system, even though the productionlevel was high, the product was almost soluble. However,in the  E. coli  system, it was over-produced as inclusion bodies. The inclusion bodies were successfully refolded andshowed approximately the same binding affinity as the WE product. These results demonstrate that using mRNAdisplay with synthetic libraries and WE and  E. coli  cell production systems, a system for in vitro selection andsmall- to large-scale production of scFvs has beenestablished. Keywords  mRNAdisplay.Antibody.scFv.Wheatembryocell-free.  E. coli  production Introduction Monoclonal antibodies (mAbs) are useful for biologicalresearch and diagnoses and have also been developed asdrugs (Fogelman et al. 2008; Hall and Cameron 2008; Onal et al. 2008). Hybridoma technology (Bussard and Pages1978; Geckeler et al. 1978; Shulman et al. 1978) has been used to produce mAbs for about 30 years. The processinvolves immunizing animals and preparing B cells to makehybridoma cells. In the past 15 years, phage display (Smith1985; Smith and Scott  1993), which can display human  proteins, has provided an alternative for the production of mAbs, mainly for therapeutic uses (Ranganathan 2008).Since the process does not require immunization and themaking of hybridoma cells, antibodies can be obtained in ashorter period of time, and moreover, by using humanantibody genes, human antibodies can easily be obtained.Recently, with advances in in vitro protein synthesis, in vitrodisplay systems, i.e., ribosome display (Hanes and Pluckthun1997) and mRNA display (Miyamoto-Sato et al. 2000;  Nemoto et al. 1997; Shibui et al. 2008; Shiratori et al. 2008), have been developed and used for engineering proteins(Fukuda et al. 2006; Lipovsek and Pluckthun 2004). The Appl Microbiol Biotechnol (2009) 84:725  –  732DOI 10.1007/s00253-009-2010-zT. Shibui ( * ) : T. Kobayashi : K. Kanatani : H. Koga :  T. IsomuraMOLECUENCE Corporation,Mitsubishi, Chemical Group Yokohama Research Center,1000 Kamoshida-cho, Aoba-ku,Yokohama, Kanagawa 227-8502, Japane-mail: MisawaMitsubishi Chemical Research Center,Mitsubishi, Chemical Group Yokohama Research Center,1000 Kamoshida-cho, Aoba-ku,Yokohama, Kanagawa 227-8502, JapanT. SasakiZOEGENE Corporation,Mitsubishi, Chemical Group Yokohama Research Center,1000 Kamoshida-cho, Aoba-ku,Yokohama, Kanagawa 227-8502, Japan  ability to make larger libraries than with in vivo displaysystems is an important characteristic of these technologies.They are able to utilize extremely large peptide libraries(more than 10 12 unique peptides; Fukuda et al. 2006). InmRNA display, protein fragments are presented at the 3 ′ terminal of their encoding mRNA through a puromycinlinker (Nemoto et al. 1997). Selection can be used to enrichclones that bind to proteins of interest, and protein sequencesof bound molecules can be decoded from their nucleotide portion. A particularly promising field of mRNA displayinvolves the construction of libraries with extensive diversityintroduced by synthetic DNAs (Shiratori et al. 2008). Withthe synthetic DNA, defined scaffolds can be used, anddiversity can be introduced in a site-specific manner (Lipovsek et al. 2007; Parker et al. 2005). Thus, high-  precision antibody engineering can be achieved usingsynthetic DNA. With synthetic libraries of sufficient diver-sity, antibodies can be engineered to 29 specifically bind toantigens through interactions with designed complementarydetermining regions (CDRs; Knappik et al. 2000). However,the general applicability of mRNA display for the screeningof synthetic human scFv libraries has not been determined.Because of the complexity and difficulties in the construc-tion of a synthetic-gene library coding for complete scFvsand instability and the formation of insoluble scFv aggre-gates in in vitro translation systems, it has been difficult toconstruct suitable synthetic human scFv libraries for mRNAdisplay.In this paper, we describe the application of mRNAdisplay technology to the generation of a synthetic scFvlibrary, successful isolation of a scFv against a protein of human origin using the library, and small- to large-scale production of this scFv in a wheat embryo cell-free systemand an  Escherichia coli  cell system. Materials and methods MaterialsThe bacterial strain, Rosetta (DE3), and its expressionvector, pET-32a, were purchased from Novagen. A vector, pGEM T Easy Vector, for sequencing was purchased fromPromega. The wheat embryo cell-free expression vector  pEU3-NII (Bardoczy et al. 2008) was a gift from Professor Endo at Ehime University. The KOD polymerase kit was purchased from Toyobo. Synthetic DNAs were obtainedfrom Operon. Synthetic DNA sequences for frameworksand CDRs of each subfamily of antibody were derived fromsequence databases (Fellouse et al. 2004; Hall and Cameron2008; Knappik et al. 2000; Tomlinson et al. 1992). To minimize the occurrence of stop codons, amino acidsequences in CDRs were encoded by semi-randomizedcodons, PQK, or in complementary strands MPQ, in whichP is T/C/A/G=10:30:30:30, Q is T/C/A/G=30:30:10:30, K is T/C/A/G=50:0:0:50, and M is T/C/A/G=0:50:50:0. Asfor the heavy-chain CDR3, semi- randomized codons inlengths of 14, 15, and 16 amino acids were used for thelibraries. Other synthetic DNAs used for PCR amplificationare listed in Table 1.Construction of V L  and V H  librariesA schematic diagram of how the V L  and V H  libraries wereconstructed for mRNA display is shown in Fig. 1a. Eachfragment was annealed, and the gap was filled in with KOD polymerase (Toyobo) according to the manufacturer  ’ s protocol except that the number of cycles in the polymerasereaction was 3. The constructed DNA fragments were usedfor mRNA display.Enrichment of complete V L  and V H  genes by mRNAdisplayThe mRNA-display molecules were produced as described previously (Shibui et al. 2008; Shiratori et al. 2008). mRNA-display molecules that contained genes encodingentire V L  and V H  fragments were enriched using Flag tagsfused at the C-terminal by beads conjugated with anti-Flagtag antibody (Sigma) similar to a method described previously (Wilson et al. 2001).Construction of a scFv library for mRNA displaywith enriched V L  and V H  librariesFigure 1 b shows a schematic diagram of how the scFvlibrary was constructed for mRNA display. In the first step,aliquots of the enriched V L  and V H  libraries were PCR-amplified three times with KOD polymerase using as primers Sp6 IRES and J/GS-linker and GS-linker/FR1andFlag tag, respectively. These amplified DNA fragmentswere combined, PCR-ligated to construct scFv DNAfragments, and re-amplified six times with KOD polymer-ase using Sp6 IRES and Flag tag. The PCR products were purified with a kit (QIAquick PCR Purification Kit,QIAGEN) and fractionated (> 200 base pairs) with a spincolumn (CHROMA SPIN-1000, BD Biosciences), and a portion was cloned into pGEM T to check their sequences.The remaining products were used for construction of ascFv library for mRNA display.Affinity selection of scFvs against human RANK To prevent nonspecific binding to antigens, the moleculesdisplaying scFvs were incubated with Protein G  –  Sepharose beads (GE Healthcare) bound to another antigen, Ephrine-Fc 726 Appl Microbiol Biotechnol (2009) 84:725  –  732  (R&D Systems), for 1 h at room temperature withrotation. The supernatant, i.e., non-bound fraction, wastransferred to human RANK   –  Fc (R&D Systems)-boundProtein G-Sepharose beads and rotated for 1 h at roomtemperature.The beads were washed three times with three-beadvolumes of the washing buffer, and then the boundmolecules were eluted two times with four volumes of 0.1 N glycin  –  HCl pH 2.5, and immediately a 0.8 volume of 1 M Tris  –  HCl pH 8.0 was added to the eluate to neutralizeit. The eluted molecules were then PCR-amplified for thenext round of selection. Subsequent rounds were performedin a similar manner.After six rounds of selection, PCR-amplified fragmentscoding for bound scFvs were cloned into pGEM T andsequenced.In vitro batch synthesis of scFvsscFv DNA fragments from the initial library, theenriched fraction after the sixth round of panning, anda sequenced clone (clone no. 10 in Figure 1 b) werePCR-amplified for 25 cycles with KOD polymeraseusing as primers, Sp6 IRES and Flag tag. mRNAs for the translation of those fragments were prepared fromthe PCR products using RiboMax large RNA Produc-tion System-SP6 (Promega). Then, scFvs were synthe-sized in 50  μ  L of translation mixture containing wheat embryo extract (A260=40, Zoegene), 20 mM Hepes/ KOH (pH 7.8), 1.2 mM ATP, 0.4 mg/mL creatinekinase, 0.64 units/  μ  L RNase inhibitor, 16 mM creatine phosphate, 0.4 mM spermidine, 100 mM potassiumacetate, 2.5 mM magnesium acetate, 2 mM DTT,0.4 mg/mL mRNA, and 0.3 mM each of 20 aminoacids in a batch manner at 26 °C for 2 h. The reactionmixtures were diluted 2-fold with PBS (10 mM sodium phosphate and 150 mM NaCl) for ELISA.ELISAAssays were conducted essentially as described (Shibuiand Nagahari 1992). Aliquots (100  μ  L) of antigensolution (3 mg/mL in PBS) were added to wells of anELISA plate (COSTAR 9018) and incubated at 4 °Covernight. The plate was washed three times with PBS.The wells were blocked with 1% BSA in PBS at 37 °C for 2 h. The plate was washed three times with 0.05% Tween20 in PBS. Samples (100  μ  L) were put into the wells andincubated for 1 h at room temperature. The plate wasagain washed three times with 0.05% Tween 20 in PBS. Next, 100  μ  L of 10 − 4 diluted anti-Flag antibody conju-gated with HRP (Sigma) was put into the wells andincubated for 1 h at room temperature.       T    a      b      l    e      1     P   r    i   m   e   r   s   u   s   e    d    f   o   r    P    C    R    P   r    i   m   e   r   s    S   u    b    f   a   m    i    l   y    S   e   q   u   e   n   c   e    (    5    '    →     3    '    )    F   o   r   c   o   n   s    t   r   u   c    t    i   o   n   o    f   s   c    F   v    D    N    A    l    i    b   r   a   r   y    f   o   r   m    R    N    A    d    i   s   p    l   a   y    S   p    6    I    R    E    S    G    G    A    A    G    A    T    C    T    A    T    T    T    A    G    G    T    G    A    C    A    C    T    A    T    A    G    A    A    C    A    A    C    A    A    C    A    A    C    A    A    C    A    A    A    C    A    A    C    A    A    C    A    A    A    A    T    G    F    l   a   g  -    t   a   g    C    C    G    C    C    C    T    T    G    T    C    A    T    C    G    T    C    A    T    C    C    T    T    A    T    A    A    T    C    A    G    A    T    G    A    T    T    T    A    T    C    G    T    C    A    T    C    G    T    C    T    T    T    G    T    A    G    T    C    C    C    C    A    C    C    A    C    J    /    G    S  -    l    i   n    k   e   r    k    1 ,    2 ,    4    C    C    A    C    T    T    C    C    A    C    C    A    C    C    A    C    C    T    G    A    T    C    C    T    C    C    T    C    C    T    C    C    A    G    C    T    T    T    A    A    T    T    T    C    C    A    C    C    T    T    G    G    T    A    C    C    C    T    G    G    C    C    G    A       λ     2 .    3    C    C    A    C    T    T    C    C    A    C    C    A    C    C    A    C    C    T    G    A    T    C    C    T    C    C    T    C    C    T    C    C    A    T    T    T    T    T    A    G    T    A    G    T    C    A    G    C    T    T    G    G    T    C    C    C    T    C    C    G    C    C    G    A    G    S  -    l    i   n    k   e   r    /    F    R    1      Η     2    A    G    G    A    G    G    A    G    G    A    T    C    A    G    G    T    G    G    T    G    G    T    G    G    A    A    G    T    G    G    T    G    G    T    G    G    A    G    G    A    T    C    T    C    A    A    G    T    T    C    A    A    C    T    G    A    A    G    G    A    G    T    C    T    G    G    G    C    C    C    G    C    T    C    T    G      Η     3    A    G    G    A    G    G    A    G    G    A    T    C    A    G    G    T    G    G    T    G    G    T    G    G    A    A    G    T    G    G    T    G    G    T    G    G    A    G    G    A    T    C    T    G    A    A    G    T    T    C    A    A    C    T    G    G    T    G    G    A    G    T    C    T    G    G    C    G    G    T    G    G    C    C    T    G      Η     6    A    G    G    A    G    G    A    G    G    A    T    C    A    G    G    T    G    G    T    G    G    T    G    G    A    A    G    T    G    G    T    G    G    T    G    G    A    G    G    A    T    C    T    C    A    A    G    T    T    C    A    A    C    T    G    C    A    G    C    A    G    T    C    T    G    G    A    C    C    C    G    G    G    C    T    G    F   o   r   e   x   p   r   e   s   s    i   o   n   e   x   p   e   r    i   m   e   n    t   s   s   c    F   v  -    F     κ     1    C    C    C    A    A    T    T    T    T    C    T    C    A    A    T    C    C    T    C    C    C    C    C    T    T    T    T    A    A    C    A    A    C    C    T    C    T    T    C    C    A    G    G    G    C    C    C    A    C    A    T    A    T    G    G    A    T    A    T    C    C    A    G    A    T    G    A    C    C    C    A    G    T    C    C    C    C    G    W    E  -    S   p    6  -    F    G    C    G    T    A    G    C    A    T    T    T    A    G    G    T    G    A    C    A    C    T    A    T    A    G    A    A    C    T    C    A    C    C    T    A    T    C    T    C    T    C    A    T    C    C    T    C    A    A    C    C    C    A    C    C    T    C    C    T    A    T    A    T    A    T    C    C    C    A    A    T    T    T    T    C    T    C    A    A    T    C    C    T    C    C    C    C    C   s   c    F   v  -    H    6  -    R    H    3    C    T    T    G    T    C    G    T    C    G    T    C    A    T    C    C    T    C    G    A    G    T    C    C    C    G    C    C    G    C    C    C    C    U    T    R  -    H    6  -    F    C    T    C    G    A    G    G    A    T    G    A    C    G    A    C    G    A    C    A    A    G    T    C    C    C    A    C    C    A    T    C    A    C    C    A    T    C    A    C    C    A    T    C    A    C    C    A    T    C    A    C    C    A    T    T    G    A    U    T    R  -    R    G    G    G    G    A    A    A    C    G    C    C    T    G    G    T    A    T    C    T    T    U    T    R  -    R  -    I   n    G    G    A    G    A    A    A    G    G    C    G    G    A    C    A    G    G    T    A    T Appl Microbiol Biotechnol (2009) 84:725  –  732 727  The plate was again washed three times with 0.05%Tween 20 in PBS. Bound scFv was detected by measuringabsorbance at 490 nm using 1.2-phenylenediamine incitrate buffer containing H 2 O 2  (Cabilly 1989).Moderate-scale cell-free production of scFv in the WEsystem with dialysisClone no. 10 was PCR-amplified from its cloned plasmidwith the primers scFv-F and scFv-H6-R. With thisamplification, a C-terminal His 6  tag sequence for affinity purification was added to the scFv sequence. 3 ′ UTR (untranslated region) was CR-amplified from pEU3-NIIusing UTR-F and UTR-R. The amplified DNA fragmentswere purified as described above and PCR-ligated usingWE-Sp6-F and UTR-R-In. mRNAs were transcribed fromthe PCR-ligated DNA fragment and purified with a kit (Promega). One milliliter of translation mixture with or without the mRNA was prepared and put in a dialysis tube(Float A Lyzer, SPECTRUM). The tube was set in a 15-mLtest tube filled with 13 mL of a feeding solution [30 mMHepes/KOH (pH 7.8), 1.2 mM ATP, 16 mM creatine phosphate, 0.4 mM spermidine, 100 mM potassium acetate,2.5 mM magnesium acetate, 4 mM DTT, and 0.3 mM eachof 20 amino acids] and incubated at 26 °C for 20 h.Construction of expression plasmid for the  E. coli  systemand productionThe PCR-amplified fragment described in the previoussection was digested with  Nde I and  Xho I and inserted between  Nde I and  Xho I sites in pET-32a (Novagen). Thesequence of the inserted fragment was confirmed, and the plasmid was introduced into Rosetta (DE3) (Novagen).Clone no. 10 was expressed according to the manufac-turer  ’ s directions. After cultivation at 30 °C overnight,cells were disrupted with BugBuster Protein ExtractionReagent (Novagen), and insoluble pellets were collected by centrifugation (20,000×  g  , 10 min.) Refolding of scFvin the insoluble fraction was done as described (Tsumotoet al. 1998).Affinity purification of clone no. 10The His 6 -tag fused clone no. 10 produced in the cell-freesystem and  E. coli  cells were partially purified with a ab Fig. 1  Scheme of the in vitroconstruction of the scFv libraryfor mRNA display.  a  Construc-tion of synthetic V L  and V H genes for mRNA display. Syn-thetic DNAs encoding theframework of human antibodyvariable regions and CDRs wereannealed, and their gaps werefilled with DNA polymerase.For enrichment of genes encod-ing full-length variable regions,mRNA-display molecules of each variable region were syn-thesized and selected with Flagtag antibody-conjugated beadsas described in  “ Materials andmethods ” .  b  Construction of thescFv library for mRNA display.The enriched VL and VHregions were re-amplified andligated with PCR. Sequences of  primers used in construction arelisted in Table 1.  Sp6   Sp6 promoter,  IRES   internal ribo-some entry sequence,  FR  anti- body framework,  CDR complementary determining re-gion,  GS-linker   Gly(Ser) 4 ×3linker 728 Appl Microbiol Biotechnol (2009) 84:725  –  732  TYRON column (Clonetech) as recommended by themanufacturer and stored at   − 80 °C prior to use.Protein concentration of the purified samplesThe concentration of the purified clone no. 10 was assayedwith a kit (Protein Assay, Bio-Rad) using BSA as astandard by following the manufacturer  ’ s protocol. In caseof the partially purified clone no. 10 produced in WE, itsscFv concentration was determined by subtracting the background level, i.e., the protein concentration of theaffinity-purified fraction without mRNA. Results Construction of a synthetic scFv library for mRNA displayWith a limited capacity to the capacity for making syntheticDNA fragments, we chose k1, k2, k4,  λ 2, and  λ 3 for frameworks of the light chain variable region (V L ) gene andH2, H3, and H6 for those of the heavy-chain variableregion (V H ) gene from germ line antibody subfamilies.Initial libraries of V L  and V H  were constructed as describedin the  “ Materials and methods ”  section. To select DNAfragments encoding complete V L  and V H  genes from theselibraries, their coding regions were mRNA-displayed, andaffinity was selected by an anti-Flag tag antibody columnusing Flag tags fused at their C-terminals.The enriched V L  and V H  genes were then PCR-ligatedand used for the construction of a scFv library for mRNAdisplay. Of the 261 independently isolated clones that weresequenced, 132 had complete and different scFv sequences.Subfamilies in the library are listed in Table 2.Selection of scFv by mRNA displayThe scFv mRNA-display library was panned against humanRANK as described in  “ Materials and methods ” . After thesixth round of panning, the nucleotide portion of theenriched-clone pool was transcribed and translated in awheat embryo cell-free system in a batch manner. As acontrol experiment, the nucleotide portion of the initialscFv library was also transcribed and translated in the samemanner. Figure 2a shows ELISA results for these pooledscFv against hRNAK and other proteins. The initial librarydid not react with hRANK; however, the scFv pool after round six specifically reacted with hRANK.Sequencing and binding to antigens of the scFvThe nucleotide portion of the enriched molecules was PCR-amplified and cloned into a plasmid vector, pGEM T, for sequencing. One sequence was detected as a redundant clone. From a database (Benson et al. 1997), its V L  and V H subfamilies were k1 and H3, respectively. Figure 3 showsalignments of the V L  and V H  portions. k1 and H3 were themost frequent subfamilies in the initial library (Table 2).One of the redundant clones was chosen (no. 10) to betranslated in the cell-free system by a batch method for ELISA.Figure 2 b shows the ELISA results. The WE reactionmixturewithoutmRNAofcloneno.10didnotbindtohIL6R,hRANK, or hIL17R; however, the reaction mixture with theclone no. 10 mRNA showed specific binding to hRANK.Moderate-scale production of clone no. 10 in the cell-freesystemTo characterize the isolated scFv more precisely, a DNAtemplate for a C-terminal His 6 -tag fusion protein was PCR-amplified from the clone no. 10 plasmid, and its proteinwas synthesized in the wheat embryo cell-free system witha dialysis method at 26 °C for 20 h as described in the “ Materials and methods ”  section. In spite of high-level production (approximately 1 mg/mL reaction mixture),most of the product was found in the soluble fraction(Fig. 4a). The proteins synthesized were affinity-purifiedwith a metal column. Co-purified proteins of about 50 kDawere observed, and they were also seen in the same fractionwithout mRNA (Fig. 4 b). The concentration of clone no. 10was calculated to be 0.14 mg/mL by the method describedin  “ Materials and methods ” .Production of clone no. 10 by  E. coli  cellsTo prepare a large quantity of clone no. 10, we chose an  E.coli  in vivo system for production.  E. coli  (ROSETTAORIGAMI) cells harboring an expression vector, pET 32ascFv, containing the same scFv-His 6  tag gene were culturedas described in the  “ Materials and methods ”  section. Cloneno. 10 was also produced well in this system; however, the product was isolated in the insoluble fraction (Fig. 5a).Theinsoluble clone no. 10 was solubilized and refolded asdescribed in the  “ Materials and methods ”  section. The Table 2  Framework analysis of the scFv DNA library for mRNAdisplayk1 k2 k4  λ 2  λ 3 Total %H2 1 0 1 2 1 5 4H3 76 6 6 22 9 119 90H6 1 1 1 5 0 8 6Total 78 7 8 29 10 132% 59 5 6 22 8Appl Microbiol Biotechnol (2009) 84:725  –  732 729
Similar documents
View more...
Search Related
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks