Protein Engineering, Vol. 12, No. 12, 1031-1034,
December 1999
© 1999 Oxford University Press
COMMUNICATIONS |
Investigation of a tetracycline-regulated phage display system
1 Institut für Biochemie (Charité), Humboldt-Universität zu Berlin, Monbijoustr.2, D-10117 Berlin and 2 Lehrstuhl für Biologische Chemie, Technische Universität München, An der Saatzucht 5, D-85350 Freising-Weihenstephan, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionReferences |
|---|
A new vector (pGZ1) was developed for bacterial phage display of antibody fragments using a transcriptional regulation element with tight control. The tetp/°-based phasmid exhibits fully suppressed scFv background synthesis in the absence of inducer and is independent of glucose as a catabolite repressor. The vector is shown to be a useful alternative to commonly used lacp/°-regulated systems.
Keywords: phagemid/pHen/single-chain Fv/tetracycline promoter
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Phage display libraries are widely used for selecting and improving properties of proteins. A lot of work has been done in the field of antibody libraries to select antibodies with desired binding specificities or catalytic activities (Janda et al., 1994
; Gao et al., 1997; Fujii et al., 1998) and for improving antibodies by affinity maturation (Gram et al., 1992; Hawkins et al., 1992; Riechmann and Weill, 1993; Low et al., 1996; Schier et al., 1996). Phage display was furthermore applied for improving the catalytic activity of enzymes by in vitro evolution (Soumillion et al., 1994; Pedersen et al., 1998).
Normally, the expression system for phage display is Escherichia coli, but foreign protein production can be toxic for this bacterium. In fact, an unfavourable selection pressure during bacterial growth is frequently observed. Bacteria not expressing the recombinant protein gain a growth advantage and can overgrow the culture. This may lead to genetic instability of the vector system (Brown and Campbell, 1993; Wülfing and Plückthun, 1993), including gene deletion or inactivation of the toxic gene product.
Working with antibody fragment-encoding libraries that represent a diverse repertoire involves both an in vitro screening for the desired characteristics of the displayed protein, such as antigen specificity, and the in vivo propagation of selected molecules. Therefore, an undesired biological selection pressure may result if the gene product is toxic for the host organism. This is generally the case if an antibody fragment is fused with gp3, the M13 minor coat protein (Krebber et al., 1997). In order to enrich proteins efficiently with the desired property, the negative biological selection pressure must be suppressed by means of stringent genetic regulation during the critical propagation steps.
Most phage display systems use vectors which are based on the lac promoter (Barbas et al., 1991; Hoogenboom et al., 1991). However, the lac promoter is leaky (Krebber et al., 1996), even if glucose is used in order to effect catabolite gene repression (Skerra, 1989). Several attempts have been made for tighter control of the lac promoter, e.g. overexpression of the lac repressor (Ørum et al., 1993) or the addition of glucose (Skerra, 1989) to improve the CAP (catabolite gene activator protein) regulated repression of the lac promoter (DeBellis and Schwartz, 1990; Hoogenboom et al., 1991). These strategies were more or less successful but could not fully suppress recombinant gene expression. The most stringently controlled lac promoter system was recently described by Krebber et al. (1996). They combined both strategies and introduced an additional upstream transcriptional terminator, thus demonstrating a reduction of background expression.
Here we propose another strategy for the safe bacterial phage display of antibody fragments using a transcriptional regulation element with reportedly tight control. The new system makes use of the tetracycline-regulated tetp/°, which was first recruited for bacterial protein production with the development of the generic expression vector pASK75 (Skerra, 1994). In form of several derivatives this vector has been successfully employed for the bacterial synthesis of Fab (Skerra, 1994), Fv and scFv (single-chain Fv) fragments (Schiweck et al., 1997). Antibody fragments were even produced at the fermenter scale without indication of plasmid instability (Schiweck and Skerra, 1995).
The advantage of this system is the very tight transcriptional regulation, which is ensured by the plasmid-encoded tet repressor, allowing convenient chemical induction by anhydrotetracycline or tetracycline. Thus, problems with host cell genotype or influence of glucose, in conjunction with catabolite repression, are avoided (Skerra, 1994). A similar type of genetic regulation was recently adopted in another expression vector (Lutz and Bujard, 1997). Recently, the tetp/° was succesfully applied in the phage display of lipocalin mutants (Beste et al., 1999).
We have now modified the vector pASK85 for expression of recombinant scFv phages in order to establish a stringently regulated phage display system, resulting in the vector pGZ1. pASK85 is a derivative of pASK75, which carries a dicistronic operon for the production of the light and heavy chains of a Fab fragment under transcriptional control of the tetp/°. For practical considerations we compare here the features of the tetracycline-regulated pGZ1 system with the conventionally used lac promoter-regulated pHen1 system (Hoogenboom et al., 1991) and we demonstrate the applicability of the tet-regulated system for phage display of antibody fragment libraries.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Vector construction
The scFv-gp3 gene was amplified in two steps from an scFv-pHen1 construct [scFv: VH-(GGGGS)3-VL] using Taq DNA polymerase (InviTek, Berlin, Germany). The first PCR (polymerase chain reaction) was performed with the primers TetB1, 5'-AACGAGGGCAAAAAATGAAATACCTATTGCCTACGG and TetF, 5'-ACAGGTCAAGCTTATTAAGACTCCTTATTACGCAGTATG. After purification of the PCR product by agarose gel electrophoresis, a second PCR was performed with the primers TetB2, 5'-CGACAAAAATCTAGATAACGAGGGCAAAAAATGAAATCAA and TetF. The amplification product was again purified by gel electrophoresis and digested with the restriction enzymes XbaI (Boehringer Mannheim) and HindIII (MBI, Fermentas). pASK85 (Skerra, 1994) was digested with the same enzymes. The gel-purified restriction fragments, the scFv-gp3 gene and pASK85 without the Fab cassette, were ligated. Supercompetent XL1-Blue cells (Stratagene) were transformed, resulting in pGZ1.
Recombinant gene expression
A 20 ml volume of 2xYT medium containing 100 µg/ml ampicillin and 1% (w/v) glucose was inoculated with 500 µl overnight culture and shaken at 37°C. Recombinant expression was induced at OD600 = 0.5 with (a) 1 mM isopropyl-ß-D-thiogalactopyranoside (IPTG) for the pHen1 construct and (b) 0.2 µg/ml anhydrotetracycline for the pGZ1 construct. The culture was shaken at 25°C. Samples were taken after 5 h. For total expression analysis 30 µl of culture were harvested by centrifugation. The pellets were resuspended in reducing SDS sample buffer and subjected to 15% SDS–PAGE, followed by Western blotting with anti c-myc antibody 9E10 (2 µg/ml). For functional scFv expression analysis, 10 ml of culture were harvested by centrifugation. The pellets were resuspended in 100 µl of periplasmic preparation buffer (Skerra, 1994), incubated for 30 min on ice and centrifuged at 14 000 rpm for 20 min. The supernatant was analyzed using scFv ELISA.
Regarding expression with different inductor concentrations, cultivation and analysis were performed as described above. Cultures were induced using different concentrations of anhydrotetracycline (0.2, 0.1, 0.05, 0.02 µg/ml) or tetracycline (20, 10, 5, 1 µg/ml).
Rescue of scFv phages
A 20 ml volume of 2xYT medium containing 100 µg/ml ampicillin and 1% (w/v) glucose was inoculated with 500 µl overnight culture of bacteria harboring scFv-pHen1 or scFv-pGZ1 and shaken at 37°C. For pHen1, at OD600 = 0.5x1011 helper phages (VCSM13, Stratagene) and 0.5 µl of 0.4 M IPTG were added and the culture was incubated at 37°C for 30 min without shaking, followed by 1 h with shaking. Then the cells were centrifuged (4000 g, 4°C) and resuspended in 50 ml of 2xYT medium containing 100 µg/ml ampicillin and 0.01 mM IPTG. The culture was shaken at 25°C. After 1 h, 25 µg/ml kanamycin was added, then the culture was shaken overnight at 25°C. For pGZ1, at OD600 = 0.5x1011 helper phages (VCSM13, Stratagene) were added and the culture was incubated at 37°C for 30 min without shaking, followed by 1 h with shaking. Then the cells were centrifuged (4000 g, 4°C) and resuspended in 50 ml of 2xYT containing 100 µg/ml ampicillin and 0.2 µg/ml anhydrotetracycline (for optimization of the scFv-gp3 display level the amount of anhydrotetracycline was changed). The culture was shaken at 25°C. After 1 h, 25 µg/ml kanamycin was added, then the culture was shaken overnight at 25°C.
The phage particles were purified and concentrated by two PEG/NaCl precipitations (Sambrook et al., 1989), resuspended in 1 ml of PBS (phosphate-buffered saline: 1.6 mM NaH2PO4, 9.2 mM Na2HPO4, 150 mM NaCl, pH 7.35), sterile filtered (0.45 µm) and stored at 4°C. The titer was measured by infecting log-phase XL1-Blue cells and counting the colonies.
Phage ELISA
NUNC ELISA plates were coated with 40 µg/ml Neutravidin (Pierce) overnight at 4°C, washed three times with PBS, 0.1% Tween (TPBS) and incubated for 30 min with 0.1 µg/ml Z-Phe-Pro-4-carboxypyridyl methyl ketone–biotin. After washing, 50 µl of phage solution (2x1010 cfu/well) and 50 µl of blocking solution (3% Gelafusal, Germed, Berlin, Germany) were applied and the mixture was incubated for 90 min at room temperature. After washing, 50 µl of horseradish peroxidase–anti-fd conjugate (Seramun, Germany) in blocking buffer was added and incubated for 90 min. Signals were developed in the presence of 50 µl of peroxidase substrate TMB (3,3',5,5'-tetramethylbenzidine) (Seramun).
scFv ELISA
Plates were coated as described above. After washing, 50 µl of periplasmic cell fraction and 50 µl of blocking solution (3% Gelafusal, Germed) were applied and the mixture was incubated for 90 min at room temperature. After washing, 50 µl of anti c-myc antibody 9E10 (2 µg/ml) in blocking buffer were added and incubated for 90 min. After another washing step, 50 µl of horseradish peroxidase–anti-mouse conjugate (Amersham Life Science) in blocking buffer were added and incubated for 90 min. Signals were developed in the presence of 50 µl of peroxidase substrate TMB.
|
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Vector construction
The phage display vector pGZ1 was derived from pASK85 (Skerra, 1994) by inserting an scFv-gp3 (scFv-A2A1, see below) expression cassette via the unique XbaI and HindIII restriction sites. The scFv-gp3 cassette was derived from pHen1 (Hoogenboom et al., 1991) using PCR for introducing the XbaI and HindIII cloning sites. pGZ1 allows the tetracycline-regulated expression of recombinant scFv as well as scFv-gp3 fusion protein in an E.coli suppressor strain such as XL1-Blue or TG1 (supE44 strains partially suppress the stop codon TAG in the reading frame between scFv and gp3). The soluble recombinant scFv can be expressed in a non-suppressor strain such as HB2151 (translation is stopped after the scFv gene). In order to analyze the properties of the new phage display vector, the anti-peptide scFv fragment A2A1 (antigen: the modified peptide Z-Phe-Pro-4-carboxypyridyl methyl ketone, a prolylendopeptidase inhibitor; Steinmetzer et al., 1993), which had been selected in our laboratory from a murine pHen1 library, was utilized. Figure 1 shows the features of pGZ1. ScFv genes can be cloned via the rare cutting SfiI and NotI sites in a generic manner. pGZ1 contains an f1 origin (from pASK85) for production of recombinant phages.
|
Characterization of the tet-regulated system
In order to characterize the new system and to compare it with the established pHen phage display system, several features were analyzed. First, we investigated the expression of recombinant phages by comparing the display level of scFv-gp3 fusion protein. Second, we analyzed the production of functional soluble scFv in the periplasmic fraction.
Recombinant phage expression
For optimization of the scFv-gp3 display level on phages, we tested different aTc (anhydrotetracycline) concentrations and analyzed the display level using phage ELISA (Figure 2). Maximum display was observed at a concentration of 0.4 µg/ml aTc. The following experiments, however, were performed with 0.2 µg/ml aTc because of the beginning of toxic effects of scFv-gp3 expression at 0.4 µg/ml (as indicated by a lower cell density).
|
For comparison of the recombinant phage expression and scFv-gp3 display level, we prepared recombinant phages with the antibody fragment A2A1 using both the pHen1 and the pGZ1 construct. The scFv-gp3 display level was analyzed using Western blotting with anti-gp3 antibody (data not shown). Functional fusion protein was detected in phage ELISA (Figure 3
). The tetp/° construct proved to be comparable to the pHen1 system with respect not only to phage titer (5x1010 to 1x1011/ml culture supernatant) but also to scFv-gp3 display level.
|
Expression analysis
Both vector systems permit the display of scFv-gp3 and recombinant expression of functional scFv. We compared the bacterial expression of the scFv A2A1 in pHen1 and in pGZ1 using XL1-Blue cells. The expression of scFv was analyzed using Western blotting and the yield of functional scFv was detected with ELISA.
Initial expression experiments revealed that the pHen construct was leaky in the absence of glucose and not fully repressed even with addition of 1% glucose (Figure 4a). The background expression under glucose is not sufficient to be detected in Western blot analysis but growth curves without induction showed a delayed growth compared with a non-expressing clone (i.e. with a frame shift in the scFv gene, data not shown). In contrast, expression from our new vector was apparently fully suppressed prior to induction (like the original pASK75 system), independent of glucose (Figure 4a), and comparison of growth curves with the non-expressing clone showed no delay in growth (data not shown).
|
In addition to the advantage of stringent repression for application in phage display selection experiments, a low level of recombinant protein expression during phage propagation is generally desirable because of the toxicity of higher amounts of gp3 fusion protein (Krebber et al., 1997
). Therefore, we tested different inducer concentrations, followed by analyzing the scFv expression. The results revealed that there is not just an on–off induction. The expression level can actually be adjusted by different anhydrotetracycline or tetracycline concentrations (Figure 4b
). Furthermore, it was found that induction with tetracycline requires a 100-fold higher amount of inducer compared with anhydotetracycline for expressing equivalent amounts of protein. ELISA analysis of soluble expression using the new construct showed a clearly higher scFv amount than when using the pHen system (Figure 5).
|
During testing of the presented system, we were able to clone a problematic scFv fragment, the murine anti-CEA scFv B4FC3 (CEA, carcinoembryonic antigen) engineered for technetium binding to be used in tumor imaging. With the lac-regulated pHen system it was not possible to clone this scFv. Even in the presence of glucose strong lysis of the colonies carrying the scFv gene was observed. All the clones with proper growth behavior resulted from frame shift escape mutations, which inactivated protein expression (Lill, 1999
). With the tet-regulated system, however, we succeeded in cloning and periplasmatically expressing this scFv gene. The scFv B4FC3 was shown to bind to CEA-coated plates in ELISA analysis comparably to the monoclonal antibody. It has to be characterized further concerning stability, binding affinity and technetium-binding properties (Küttner G., personal communication). Conclusions
The tetp/°-based phasmid construct described here for use in antibody phage display libraries has a fully suppressed scFv background synthesis in the absence of inducer and it is independent of glucose. The vector was demonstrated to be a useful alternative to the ordinary lac promoter-regulated systems, with comparable properties concerning phage production, scFv-gp3 fusion protein expression or soluble scFv expression after induction. The tight transcriptional control of this system should, however, provide a clear advantage in the functional screening of highly diverse antibody fragment libraries and in the cloning of toxic antibody fragments.
|
Acknowledgments |
|---|
We thank K.Winkler and M.Paschke for helpful discussions. Part of the work was supported by the Deutsche Forschungsgemeinschaft (grant INK 16/A1-1).
|
Notes |
|---|
3 To whom correspondence should be addressed
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Barbas,C.F., Kang,A.S., Lerner,R.A. and Benkovic,S.J. (1991) Proc. Natl Acad. Sci. USA, 88, 7978–82.
Beste,G., Schmidt,F.S., Stibora,T. and Skerra,A. (1999) Proc. Natl Acad. Sci. USA, 96, 1898–1903.
Brown,W.C. and Campbell,J.L. (1993) Gene, 127, 99–103.[ISI][Medline]
DeBellis,D. and Schwartz,I. (1990) Nucleic Acids Res., 18, 1311.
Fujii,I., Fukuyama,S., Iwabuchi,Y. and Tanimura,R. (1998) Nature Biotechnol., 16, 463–467.[ISI][Medline]
Gao,C., Lin,C.H., Lo,C.H.L., Mao,S., Wirsching,P., Lerner,R.A. and Janda,K.D. (1997) Proc. Natl Acad. Sci. USA, 94, 11777–11782.
Gram,H., Marconi,L.A., Barbas,C.F., Collet,T.A., Lerner,R.A. and Kang,A.S. (1992) Proc. Natl Acad. Sci. USA, 89, 3576–80.
Hawkins,R.E., Russell,S.J. and Winter,G. (1992) J. Mol. Biol., 226, 889–96.[ISI][Medline]
Hoogenboom,H.R., Griffiths,A.D., Johnson,K.S., Chiswell,D.J., Hudson,P. and Winter,G. (1991) Nucleic Acids Res., 19, 4133–4137.
Janda,K.D., Lo,C.H., Li,T., Barbas,C. F3., Wirsching,P. and Lerner,R.A. (1994) Proc. Natl Acad. Sci. USA, 91, 2532–2536.
Krebber,A., Burmester,J. and Plückthun,A. (1996) Gene, 178, 71–74.[ISI][Medline]
Krebber,A., Bornhauser,S., Burmester,J., Honegger,A., Willuda,J., Bosshard,H.R. and Plückthun,A. (1997) J. Immunol. Methods, 201, 35–55.[ISI][Medline]
Lill,C. (1999) Dissertation, Humboldt University, Berlin.
Low,N.M., Holliger,P.H. and Winter,G. (1996) J. Mol. Biol., 260, 359–68.[ISI][Medline]
Lutz,R. and Bujard,H. (1997) Nucleic Acids Res., 25, 1203–1210.
Ørum,H. andersen,P.S., Oster,A., Johansen,L.K., Riise,E., Bjornvad,M., Svendsen,I. and Engberg,J. (1993) Nucleic Acids Res., 21, 4491–4498.
Pedersen,H., Holder,S., Sutherlin,D.P., Schwitter,U., King,D.S. and Schultz,P.G. (1998) Proc. Natl Acad. Sci. USA, 95, 10523–10528.
Riechmann,L. and Weill,M. (1993) Biochemistry, 32, 8848–55.[Medline]
Sambrook,J., Fritsch,E.F. and Maniatis,T. (1989) Molecular Cloning. A Laboratory Manual. 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Schier,R., Bye,J., Apell,G., McCall,A., Adams,G.P., Malmqvist,M., Weiner,L.M. and Marks,J.D. (1996) J. Mol. Biol., 255, 28–43.[ISI][Medline]
Schiweck,W. and Skerra,A. (1995) Proteins, 23, 561–565.[ISI][Medline]
Schiweck,W., Buxbaum,B., Schätzlein,C., Neiss,H.G. and Skerra,A. (1997) FEBS Lett., 414, 33–38.[ISI][Medline]
Skerra,A. (1989) Dissertation, Ludwig Maximilians University, Munich.
Skerra,A. (1994) Gene, 151, 131–135.[ISI][Medline]
Soumillion,P., Jespers,L., Bouchet,M., Marchand Brynaert,J., Winter,G. and Fastrez,J. (1994) J. Mol. Biol., 237, 415–422.[ISI][Medline]
Steinmetzer,T., Silberring,J., Mrestani,K.C., Fittkau,S., Barth,A. and Demuth,H.U. (1993) J. Enzym. Inhib., 7, 77–85.[ISI][Medline]
Wülfing,C. and Plückthun,A. (1993) Gene, 136, 199–203.[ISI][Medline]
Received April 6, 1999; revised August 20, 1999; accepted August 23, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. H. Bessette, J. J. Rice, and P. S. Daugherty Rapid isolation of high-affinity protein binding peptides using bacterial display Protein Eng. Des. Sel., October 1, 2004; 17(10): 731 - 739. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||