Expression of anti-Z-DNA single chain antibody variable fragment on the filamentous phage surface

Braz J Med Biol Res, May 2000, Volume 33(5) 569-579

Expression of anti-Z-DNA single chain antibody variable fragment on the filamentous phage surface

A.Q. Maranhão and M.M. Brígido

Departamento de Biologia Celular, Instituto de Biologia, Universidade de Brasília, Brasília, DF, Brasil

^{Correspondence and Footnotes}

We describe the expression of an anti-Z-DNA single chain variable region antibody fragment (scFv) on a filamentous phage surface. Four vectors for phage display were constructed. Two of them are able to display multiple copies of the antibody fragment, and the others can be used to make monovalent libraries. The vectors use different promoter/leader sequences to direct the expression of the fused proteins. All were able to promote the assembly of fusion virion particles. In this paper we also show the affinity selection (biopanning) of those phage-antibodies based on the capacity of their products to recognize the antigen. We used biotinylated Z-DNA and the selection was performed in a solution phase fashion. The data presented here indicate that these vectors can be further used to construct anti-nucleic acid antibody fragment libraries that can be used to study the basis of nucleic acid-protein interaction and its role in autoimmunity mechanisms.

Key words: anti-nucleic acid antibodies, phage display, biopanning, scFv, Z-DNA

Abstract

Introduction

Anti-Z-DNA antibodies are found in serum of patients with autoimmune diseases such as systemic lupus erythematosus and rheumatoid arthritis (1). It is also possible to induce them by immunization with Br-(dG-dC)_n, a stable Z-DNA conformation polymer (2). Thus, the study of the basis of nucleic acid recognition by this kind of antibodies can provide insights into the structural relationship between autoantibodies and vaccinal antibodies to this class of antigens. These studies must include the production of mutant forms of those antibodies and measurement of their residual affinity/specificity.

An important technology that arose in 1988 for studying macromolecular interactions is the display of foreign proteins and peptides on the surface of filamentous phages (3). This technique permits the generation of libraries of molecules with affinities for certain ligands, and when it is associated with biopanning and selection, new high-affinity forms can be isolated. Antibodies (particularly their Fvs - variable fragment and/or Fabs - antigen binding fragments) are the most common biomolecules expressed on the filamentous phage surface. The association of phage display libraries with biopanning can mimic the immune system clonal selection (4). The development of the antibody display on the filamentous surface has been the subject of extensive investigations (for a more extensive review, see Ref. 5). Using this approach it was possible to isolate new high-affinity antibody forms (4,6), or even change specifities of well-characterized ones (7,8).

Phage display libraries are usually assembled in two different kinds of vectors. The initial studies were done using phage vectors derived from the filamentous phage genome, such as fd, f1 and M13 (9). More recently, the literature has proposed the use of phagemid vectors containing the filamentous phage intergenic region that comprises the rolling circle phage origin of replication which enables it to pack into the fusion virion particle (10). This last type of vector is used to transform the F⁺ host strain and the library can be rescued by helper phage (11). The advantage of using phagemids is the easier manipulation of a plasmid DNA over a phage genome that facilitates cloning and DNA preparation.

A phagemid to be used for phage display library purposes must have some specific features. First, it should be able to produce N-terminal fusion molecules (library) with one of the phage coat protein. Usually, the most common are protein VIII, the major coat protein (pVIII), and protein III, one of the proximal end protein responsible for virus infectivity (pIII). For this purpose, a leader sequence (for secretion) and a weak promoter always precede the fusion system. Besides these characteristics, a phagemid must have a selection marker (to allow selection of the transformed strains) and both bacterial and viral replication origins (12).

The choice of gene VIII or III fusion is based on the experimental hypothesis addressed. Since gene VIII expresses the major coat protein, the fusion molecule with protein VIII is displayed on the virion surface in multiple copies, so the selection is also a function of the avidity of these repeated molecules. These libraries are called poly- or multivalent (13). If the choice is to construct a monovalent library, the fusion virion partner should be protein III since each virion has 5 to 6 copies of them (14) and the fusion particle would display 1 to 3 fusion molecules (13). Recent studies have proposed the use of multivalent libraries for initial screening experiments and in a second step, when refined ligands are desired, the monovalent libraries would be the choice (15).

In this paper we describe the display of an anti-Z-DNA antibody fragment (scFv) on the filamentous bacteriophage surface in a mono- or multivalent fashion. To produce these fusion particles we have constructed two new vectors and modified a third one to construct two additional ones. We also show that the particles that display anti-Z-DNA are selected by the interaction with biotinylated Z-DNA. This study represents the first step in identifying the amino acid residues that play an important role in the nucleic acid recognition by antibodies.

Material and Methods

Construction of pAIg 316 and 816 vectors

Multistep construction was carried out using standard molecular biology techniques (16). pAIg 316 was constructed in two steps: initially, the M13 gene III was amplified by PCR using a pair of primers (5' GCCCATGGCTCCGGTACCGGTACCGAAACTGTTGAAAGTTG 3' and 5' GCGAATTCTGGCATGATTAAGACTC 3') designed to generate a fragment of 1.2 kb, encoding the mature form of pIII and flanked by NcoI and EcoRI restriction sites. The design was performed based on M13 mp18 sequence data obtained from the gene bank (access number X02513). The PCR product was digested with both restriction endonucleases and cloned into pIg 16, a vector that codes for the scFv fragment of the anti-Z-DNA Z22 antibody (17). This vector was named pIg 316. The next step was transfer the fusion (scFv/gene III) to the pGEM 3Z (f-) phagemid (Promega^®, Madison, WI, USA). The recipient plasmid and the donor were digested with XmaI and EcoRI, and the fusion fragment was ligated to pGEM 3z phagemid, giving origin to phagemid pGIg 316. The protein A promoter and leader sequence (PPLA) was initially obtained from the pGTT 15 plasmid (18) by PCR and subsequently digested with BclI and HindIII endonucleases. The sequences of the primers used were 5' GATTTAGGTGACACTATAG 3' and 5' CGCCCGGGCTTTTGTCACAGG 3' for the amplification of the promoter/leader sequence from upstream sequences and carboxyl terminus, respectively. The PPLA fragment was cloned into pGIg 316 BamHI and HindIII unique sites, giving origin to phagemid pAIg 316, which is able to express Z22 scFv fused with pIII. Phagemid pAIg 816 was constructed by replacing the pAIg 316 gene III with gene VIII. Gene VIII was obtained by PCR, using a pair of primers (5' CCCATGGCTCCGGTACCGCTGAGGGTGACGAT 3' and 5' GAGCCTTGAATTCTATCGGTTTATC 3') designed to amplify the mature portion of the M13 pVIII gene (approximately 150 bp), flanked by NcoI and EcoRI restriction sites. The primers were designed based on the same sequence as described above (access number X02513). The PCR product was ligated to the pAIg 316 vector gene III-less fragment, and the resulting phagemid was called pAIg 816. The two vectors were checked by nucleotide sequencing of the fusion regions: leader sequence/scFv and scFv/gene III and VIII.

Construction of pCIg 316 and 816 vectors

Phagemids named pCIg were derived from the pCANTAB 5E Pharmacia^® vector (Uppsala, Sweden). To construct these vectors the linear recipient plasmid pCANTAB 5E was ligated to a DNA linker, harboring XmaI, BglII, XbaI and NcoI restriction sites, giving rise to the pCL vector. The oligonucleotides used to make the synthetic linker showed the following sequences: 5' CGGCCCGGGAAGATCTCTAGATCCCATGGTGC 3' and 5' GGCCGCACCATGGGATCTAGAGATCTTCCCGGGCCG 3'. The annealing procedure was performed by mixing equimolar quantities of both oligonucleotides, heating them and allowing them to cool down to room temperature under high salt conditions. The annealed linker harbors cohesive ends (NotI and SfiI) that permit it to ligate to linearized pCANTAB 5E. The pCL vector was digested with XmaI and NcoI and ligated to the scFv fragment isolated from pGIg 316, giving origin to pCIg 316. Phagemid pCIg 816 was obtained by replacing the pCIg 316 gene III with gene VIII (the same used to construct pAIg 816), that was cloned into the pGEM-T vector, showing NcoI and EcoRI restriction sites. As a negative control for fusion protein production, the pCIg 16 plasmid was also constructed from pCL, introducing the fusion scFv/protein A from the pIg 16 vector instead of scFv/gene III or VIII at the same sites, XmaI and EcoRI. This last vector (pCIg 16) produces soluble Z22 scFv without a phage protein partner. The constructions were checked by nucleotide sequencing, mainly in their fusion regions: scFv/gene III or VIII.

Phage production and panning

Fusion phage particles were produced, titrated and selected using a protocol modified from the Recombinant Phage Selection Module (Pharmacia^®). Briefly, transformed E. coli cells (strain TG1) were inoculated into 2X YT medium containing 2% glucose. The cultures were incubated at 37^oC with shaking at 250 rpm until they reached an A₆₀₀ of 0.5. At this point ampicillin was added to a final concentration of 100 µg/ml and 3 x 10¹⁰ plaque-forming units of helper phage M13K07. The cultures were incubated for 1 h under the same conditions and then spun at 3,000 g for 10 min to sediment the cells. The cell pellets were resuspended in 2X YT medium containing ampicillin (100 µg/ml) and kanamycin (50 µg/ml) and incubated overnight at 37^oC with shaking at 250 rpm. The supernatants, containing fusion phages, were obtained by centrifuging the cell cultures at 3,000 g for 20 min. The phage preparations were titrated to determine the yield of ampicillin-transducing units per milliliter for each construction. These procedures were carried out by infecting log-phase E. coli TG1 cells with serial supernatant dilutions. A typical phage yield obtained was 10¹⁰ to 10¹¹ ampicillin-transducing units/ml. The amount of transducing units used for solution phase selection was 3 to 4 x 10¹⁰ for each construction. In this experiment we used transducing units from all five constructions: pAIg 316, pAIg 816, pCIg 316, pCIg 816 and also pCIg 16 (virion particle without fusion protein) as a negative control.

The selection procedure was performed in a final volume of 1.4 ml containing the phage suspension diluted in PBS, blocking buffer (280 µl of 10% nonfat milk) and biotinylated Z-DNA (7 µl of a stock solution of 66 µg/ml prepared as described in Ref. 17). Initially, the fusion phage suspension was incubated with blocking buffer for 10 min followed by incubation with the antigen for 1 h at room temperature, with gentle rocking on a shaker table. At this point 60 µl of resuspended streptavidin-agarose resin was added and further incubated for 20 to 60 min under the same conditions. The samples were spun at 1,000 g for 1 min and the supernatant was removed. The pellet was washed 6 times with 0.05% PBS-Tween 20 by discarding the supernatant after centrifugation at 1,000 g. The washed agarose was transferred to 10 ml of log-phase TG1 E. coli cells to allow infection. Several dilutions of the transduced cells were plated onto SOBAG/ampicillin plates and incubated at 30^oC, and the number of individual clones was determined. After this first round of selection, additional rounds were also performed. The estimated amount of transducing units per cycle was determined for each construction by subtracting the average number of colonies obtained with transformed pCIg 16 cells. We used the average number of pCIg 16 transducing units to represent the background of nonspecific binding of helper phage particles.

To produce virion particles harboring Z22 scFv we constructed two new vectors and modified the Pharmacia pCANTAB 5E vector, creating two new versions of the latter.

Vector pAIg 316 was assembled by amplifying and cloning both virus gene (gene III) and promoter and leader sequences of the Staphylococcus aureus protein A gene. Gene III was obtained from the M13 genome with the primers described in Material and Methods. These primers create restriction enzyme sites (NcoI at the 5' end and EcoRI at the 3' end). The 1.2-kb fragment was digested with both enzymes and cloned into the pIg 16 vector (17). The whole cassette (scFv/gene III) was then transferred to the Promega pGEM 3Z f(-) vector in order to receive the promoter and leader sequences of the Staphylococcus aureus protein A gene (fragment PPLA). This new vector (pGIg 316) was then digested with HindIII and BamHI to receive the PPLA fragment. This last fragment was obtained by PCR of the S. aureus protein A gene cloned into the pGTT 15 vector (18). The PCR fragment was used instead of the original pGTT 15 to avoid the inhibitory effect of methylation of the BclI site. The PCR fragment (1.5 kb) was then cleaved with HindIII and BclI, giving origin to a 0.8-kb fragment that was purified and cloned into linear pGIg 316. This vector was named pAIg 316 (the whole strategy of construction of this vector is presented in Figure 1). The production of the fusion protein was observed in Southern-Western blot experiments by comparing the ability of the fusion and wild type particles to bind Z-DNA (data not shown).

Figure 1
- Schematic representation of pAIg vector constructions. The M13 gene III was obtained by PCR with specially designed oligonucleotides (see Material and Methods) and then transferred to the pIg 16 vector (17). This originated the pIg 316 vector. The whole cassette single chain variable region antibody fragment (scFv)/gene III was cloned into XmaI and EcoRI sites of the Promega vector pGEM 3Z (f-), giving rise to the pGIg 316 plasmid. The fragment containing the S. aureus protein A promoter and leader sequence (PPLA) was obtained by PCR and a subsequent digestion with BclI and HindIII from the pGTT 15 vector (18). This insert was cloned into HindIII and BamHI sites of pGIg 316, originating the pAIg 316 phagemid. To construct pAIg 816, the M13 gene III was replaced by the M13 gene VIII, which was obtained by PCR with specially designed primers from the M13 genome. Gene VIII was cloned into NcoI and EcoRI sites.

To construct the pAIg 816 vector (the strategy is also presented in Figure 1), the M13 gene VIII was amplified by PCR and cloned between the NcoI and EcoRI sites of the gene III-less pAIg 316 phagemid.

The phagemids pCIg 316 and 816 were constructed from the pCANTAB 5E vector (Pharmacia). This plasmid is available commercially in linear form. To transfer Z22 scFv to this vector we constructed and cloned a synthetic linker and the resulting plasmid was called PCL. Between XmaI and NcoI PCL restriction sites we cloned the pIg 16 scFv, and this vector was called pCIg 316. Phagemid pCIg 816 was obtained by replacing pCIg 316 fd gene III with M13 gene VIII (obtained by PCR as described for pAIg 816). The strategies of these constructions are schematically presented in Figure 2. As a negative control, we also constructed pCIg 16, which is a vector that produces soluble scFv, without a virus partner. Thus, cells harboring pCIg 16 produce wild type phage particles.

The phagemids express a pIII (316) or pVIII (816) product fused with anti-Z-DNA scFv. In two of them (pAIgs) the production of fusion molecules is under the control of the Staphylococcus aureus protein A promoter and leader sequence, while the other two (pCIgs) are derived from the pCANTAB 5E Pharmacia^® vector (which has the pLac-lactose operon promoter, and the fd phage gene III leader sequence).

The open reading frame-predicted sequences of the fusion proteins are shown in Figure 3. The nucleotide sequences of all four vector fusion regions were determined by sequencing procedures and are highlighted in the same figure. These determinations revealed that at least at these regions the open reading frames, as expected, were able to produce fusion proteins in a correct frame.

Figure 2
- Schematic representation of pCIg vector constructions. The Pharmacia pCANTAB 5E vector was modified to express the Z22 single chain variable region antibody fragment (scFv) fusioned with fd gene III. This commercial vector is available in a linear fashion. To clone the scFv, we designed self-annealed oligonucleotides to create a linker (see Material and Methods). The annealed oligonucleotides were cloned into the pCANTAB 5E vector, giving rise to the pCL plasmid. The scFv was obtained from the pIg 16 vector (17). This cloning procedure originated the pCIg 316 vector. To construct pAIg 816, fd gene III was replaced by M13 gene VIII, which was obtained by PCR with specially designed primers from the M13 genome. Gene VIII was cloned into the NcoI and EcoRI sites, giving rise to the pCIg 816 phagemid.

Figure 3
- Open reading frame and predicted fusioned polypeptides coded by phage display vectors. A, Open reading frame and amino acid prediction of scFv-M13 gene VIII fusion cloned into pAIg 816 vector. The non-coding sequence of PPLA fragment is in lower case. The restriction sites used for the cloning procedure are shown. The underlined nucleotides were checked by sequencing procedures. These regions were also sequenced for the other three vectors. B, Open reading frame of M13 gene III amplified by PCR from the M13 genome. The introduced sites are shown. C, Sequence of the pLac and fd gene III leader sequence of pCIg vectors. The introduced XmaI site is shown. This restriction site was used to clone Z22 anti-Z-DNA scFv. The non-coding region of pLac is in lower case.

Fusion virion particles were produced by standard protocols and selected in a solution-phase assay: phage antibodies were mixed with biotinylated antigen (Z-DNA) and the complex was captured from solution with streptavidin-agarose resin. The washed streptavidin/biotinylated antigen/phage antibody complex was used to infect log-phase E. coli TG1 cells which could be rescued with the helper phage, selected again or plated onto SOBAG plates. All four vectors were able to produce phage-displaying scFv, since it was possible to select them using Z-DNA as antigen.

The results of three rounds of selection (Table 1) show a 100- to 10,000-fold increase in transducing units/ml in round 3 when compared with round 1 of selection. This is expected, since no diversity was introduced in the scFv sequence from one round to another, so that selection favored the virion particles able to bind to Z-DNA.

 

Using these four vectors one would expect a larger amount of colonies arising from 816 vectors than from 316 vectors due to their multivalent display of fusion proteins (19). According to our results, only pCIgs showed this feature: in round 3 we obtained about 20 times more transducing units/ml using pCIg 816 than using pCIg 316. The data for the pAIg vectors did not show this behavior; in fact pAIg 316 yielded 5 times more transducing units/ml than pAIg 816. This discrepancy may be the result of the expression level driven by two different promoters. The staphylococcal promoter (pAIg vectors) is a constitutive promoter while pLac (pCIg vectors) is an inducible one. In our experimental procedure for phage production we used a condition of no induction of pLac (medium without inductor), so that the expression of scFv/pIII or pVIII is a result of basal pLac expression, and, in this situation, the expression is considered to be weak (20). On the other hand, Escherichia coli cells harboring the original protein A expression vector, pGTT 15, express a constitutively significant amount of this staphylococcal protein (data not shown). Perhaps the expression driven by this promoter/leader sequence is high enough to affect virus production (21). Another aspect that should be considered is that our scFv-phage was selected against Z-DNA, which is a monotonous molecule, with repeated epitopes (22). This makes this antigen multivalent itself, so that a single Z-DNA molecule could bind to more than one phage, masking a multivalence library effect.

The vectors and protocols presented in this paper suggest a new experimental model for displaying libraries of anti-Z-DNA scFv. These vectors may help to investigate the basis of DNA-protein interaction and could be useful to generate specificities for nucleic acid such as those found in patients with autoimmune disease.

Results and Discussion

References

1. Polymenis M & Stollar BD (1994). Critical binding site amino acid residues of anti-Z-DNA single chain Fv molecules: the role of H and L chain CDR3 and relationship to autoantibody activity. Journal of Immunology, 152: 5318-5329.

2. Brigido MM & Stollar BD (1991). Two induced anti-Z-DNA monoclonal antibodies use VH gene segments related to those of anti-DNA autoantibodies. Journal of Immunology, 146: 2005-2009.

3. Parmley SF & Smith GP (1988). Antibody selectable filamentous fd phage vectors: affinity purification of target genes. Gene, 73: 305-318.

4. Hoogenboom HR & Winter G (1992). By-passing immunisation - human antibodies from synthetic repertoires of germline V_H gene segments rearranged in vitro. Journal of Molecular Biology, 227: 381-388.

5. Hoogenboom HR, de Bruïne AP, Hufton SE, Hoet RM, Arends JW & Roovers RC (1998). Antibody phage display technology and its applications. Immunotechnology, 4: 1-20.

6. Hawkins RE, Russel SR & Winter G (1992). Selection of phage antibodies of a structural epitope by using a peptide library displayed on filamentous bacteriophage. Journal of Immunology, 153: 724-729.

7. Garrad LJ & Henner DJ (1993). Selection of an anti-IGF-1 Fab from a Fab phage display library created by mutagenesis of multiple CDR loops. Gene, 128: 103-109.

8. Ward ES (1995). VH shuffling can be used to convert a Fv fragment of anti-hen egg lysozyme specificity to one that recognizes a T cell receptor Va. Molecular Immunology, 32: 147-156.

9. Smith GP (1993). Surface display and peptide libraries. Gene, 128: 1-2.

10. Lowman HB (1997). Bacteriophage display and discovery of peptides leads for drug development. Annual Review of Biophysics and Biomolecular Structure, 26: 401-424.

11. Smith GP & Scott JK (1993). Libraries of peptides and proteins displayed on filamentous phage. Methods in Enzymology, 217: 228-257.

12. Breitling F, Dübel S, Seehaus T, Fuchs P, Braunagel M, Klewinghaus I & Little M (1991). A surface expression vector for antibody screening. Gene, 104: 147-153.

13. Lowman HB, Bass SH, Simpson N & Wells JA (1991). Selecting high-affinity binding proteins by monovalent phage display. Biochemistry, 30: 10832-10838.

14. Makowski L (1993). Structural constraints of the display of foreign peptides on filamentous bacteriophages. Gene, 128: 5-11.

15. Wells JA (1996). Hormone mimicry. Science, 273: 449-450.

16. Sambrook J, Fritsch EF & Maniatis T (1989). Molecular Cloning - A Laboratory Manual. 2nd edn. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

17. Brígido MM, Polymenis M & Stollar BD (1993). The role of mouse V_H10 and V_L gene segments in the specific binding of antibody to Z-DNA, analyzed with recombinant single chain Fv molecules. Journal of Immunology, 150: 469-475.

18. Brígido MM, Baradi CRM, Bonjardin CA, Santos CLS, Junqueira ML & Brentani RR (1991). Nucleotide sequence of a variant protein A of Staphylococcus aureus suggests molecular diversity among strains. Journal of Basic Microbiology, 31: 337-345.

19. Whringhton NC, Farrell FX, Chang R, Kashyap AK, Barbone FP, Mulcahy LS, Johnson DL, Barrett RW, Jolliffe LK & Dower WJ (1996). Small peptides as potent mimetic of the protein hormone erythropoietin. Science, 273: 458-463.

20. Younderian P (1988). Promoter strength: more is less. Trends in Genetics, 4: 327-328.

21. Posner B, Smiley J, Lee I & Benkovic S (1994). Catalytic antibodies: perusing combinatorial libraries. Trends in Biochemical Sciences, 19: 145-150.

22. Stollar BD (1992). Immunochemical analyses of nucleic acids. Progress in Nucleic Acid Research and Molecular Biology, 49: 39-77.

Acknowledgments

We wish to thank Cynthia Maria Kyaw for helping with the figures and Linda S. Caldas and Lídia Maria Pepe de Moraes for revising the English text.

Address for correspondence: M.M. Brígido, Laboratório de Biologia Molecular, Departamento de Biologia Celular, Instituto de Biologia, UnB, Campus Universitário, Asa Norte, 70910-900 Brasília, DF, Brasil. Fax: [phone]. E-mail: [e-mail]

Research supported by FAP-DF (Fundação de Apoio à Pesquisa do Distrito Federal). Received July 28, 1999. Accepted February 4, 2000.

Correspondence and Footnotes

Publication Dates

• Publication in this collection
  20 Apr 2000
• Date of issue
  May 2000

History

• Accepted
  04 Feb 2000
• Received
  28 July 1999