Previous Article | Next Article ![]()
Applied and Environmental Microbiology, November 2004, p. 6936-6939, Vol. 70, No. 11
0099-2240/04/$08.00+0 DOI: 10.1128/AEM.70.11.6936-6939.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
| SHORT REPORT |
Departamento de Biotecnología, Instituto de Agroquímica y Tecnología de Alimentos (CSIC), Burjassot,1 Departamento de Microbiología, Facultad de Medicina, Universitat de València, Valencia, Spain2
Received 26 February 2004/ Accepted 12 July 2004
ABSTRACT
Single-chain antibodies (scFv) recognizing the VP8* fraction of rotavirus outer capsid and blocking rotavirus infection in vitro were isolated by phage display. Vectors for the extracellular expression in Lactobacillus casei of one of the scFv were constructed. L. casei was able to secrete active scFv to the growth medium, showing the potential of probiotic bacteria to be engineered to express molecules suitable for in vivo antirotavirus therapies.
Group A rotaviruses are the main cause of diarrhea in children worldwide and are estimated to be responsible for more than 800,000 deaths of children under 5 years of age each year (12). The incidence of rotavirus disease in developing countries as well as the lack of an effective vaccine supports the development of new and more effective antirotaviral strategies. Mucosal immunity against rotavirus infection is believed to rely mainly on the production of rotavirus-specific immunoglobulin A (IgA) antibodies at the intestinal mucosal surface (12, 17, 26). Moreover, lacteal IgG or monoclonal IgA against the VP8* portion of rotaviral VP4 outer capsid can protect newborn mice against rotavirus-induced diarrhea (5, 21, 22). This ability makes VP8* a good target for the design of antirotavirus therapies.
Lactic acid bacteria (LAB), which are generally recognized as safe by the U.S. Food and Drug Administration, are microorganisms that are present in numerous food fermentations and are also normal constituents of the intestinal habitat. In addition, some strains of LAB exhibit probiotic properties. These characteristics have been exploited for the use of LAB as live vectors for the expression of different peptides and the delivery of peptides to mucosal surfaces in animal models. These peptides include antigens (16), interleukins (24), enzymes (3), and single-chain antibodies (scFv) (1, 13). These last molecules are chimeric proteins consisting of a fusion of the variable heavy (VH) and variable light (VL) regions of immunoglobulins (19). Specific scFv can be isolated by the phage display technique after panning of phage scFv libraries on immobilized antigen (7, 9). scFv offer very interesting clinical perspectives; although they may not be as powerful as natural immunoglobulins, many possible applications can be envisaged, since scFv can be cloned, manipulated, and produced in microbial hosts (20). Two cases of successful therapeutic application by in vivo delivery of scFv in mucosae by LAB have been reported (1, 13). We decided to construct Lactobacillus casei strains expressing extracellular or cell-wall-attached anti-VP8* scFv, which might be useful to deliver passive immunity against rotavirus. The use of L. casei might be of particular interest, since some strains exhibit an intrinsic beneficial effect in the treatment of rotaviral diarrhea (11, 18).
Isolation of scFv against VP8*.
The Griffin.1 phage display library (6) was used to select phage antibodies against purified VP8* from the rotaviral SA11 strain. This library is a semisynthetic human scFv library composed of more than 109 independent clones carrying VH and VL immunoglobulin variable regions cloned into pHEN2 (8) to produce an scFv fused to the pIII protein of the M13 viral capsid. Several rounds of panning and selection of VP8*-binding phages were carried out as described previously (9) with VP8*-coated immunotubes (Polysorp; NUNC). Titers of eluted phages and their signals in VP8*-specific enzyme-linked immunosorbent assay (ELISA) increased after each round, indicating the enrichment of VP8*-specific phages (data not shown). Phages were rescued from several individual clones from the third and fourth rounds of selection and tested for their ability to bind VP8* by ELISA. From 96 assayed clones, 65 phages turned out to be positive, showing signals ranging from weakly (A414 = 0.2) to strongly (A414 > 1.5) positive. Fifteen independent clones representing both weak and strong positives were randomly chosen and assayed in a second ELISA. As can be seen in Fig. 1A, phages derived from all of the clones produced a positive ELISA with VP8* but not with the bovine serum albumin (BSA) negative control.
![]() View larger version (46K): [in a new window] |
FIG. 1. Isolation of anti-VP8* scFv by phage display. (A) ELISA analysis of clones recognizing VP8* and isolated from rounds 3 and 4. BSA was used as the negative control. (B) Western blot analysis of 50 ng of VP8* probed with 107 M13 phages/ml carrying anti-VP8* scFv::pIII fusions. The blots were developed with HRPO-conjugated anti-M13 (Amersham).
|
Positive clones were assayed for recognition of VP8* by Western blot analysis using the whole M13 phages as an antibody reagent. Fifty nanograms of VP8* was subject to sodium dodecyl sulfate-polyacrylamide gel electrophoresis, electrotransferred to nitrocellulose membranes, and probed with M13 phages (107 PFU/ml in phosphate-buffered saline [PBS] plus 2% skim milk). Blots were developed with an anti-M13 antibody conjugated to alkaline phosphatase (Amersham) and nitroblue tetrazolium-BCIP (5-bromo-4-chloro-3-indoylphosphate) (Roche) as the substrate (Fig. 1B). Only clones 2B3, 2E4, and 2F6 performed well in Western blot analysis and were able to detect denatured VP8* transferred to nitrocellulose membranes, indicating the recognition of linear epitopes in VP8*. Clones were subjected to PCR amplification of the scFv coding gene with oligonucleotides FOR_linkseq (5'-GCCACCTCCGCCTGAACC-3') and pHEN-SEQ (5'-CTATGCGGCCCCATTCA-3') and sequenced. Inspection of the complementarity-determining region 3 (CDR3) of VH revealed that the 15 clones carried eight different sets of sequences (Table 1). The clones exhibited high variability in their sequences, although four clones (2A1, 2B3, 2E3, and 2E11) had a biased amino acid composition in their CDR3, with preference for the basic amino acid arginine and for the presence of tryptophan. However, at this stage it cannot be stated whether scFv sharing homologous CDR3 are recognizing similar epitopes in VP8*. Surprisingly, one clone (2H6) was shown to have originated from a pHEN2 clone of the Griffin.1 library, which did not receive the VL region during cloning, and it thus contained an in-frame fusion between the signal peptide carried by pHEN2, the VH region, the linker region, and the pIII-coding gene. Clones 2E4 and 2F6, which were shown to be functional in Western blot analyses, were identical in sequence. The third clone that was positive by Western blot analysis, 2B3, shared the same CDR3 sequence as clone 2E11, although the two clones differed in their VL sets of sequences (Table 1), which might explain the inability of clone 2E11 to recognize denatured VP8* on a membrane.
|
View this table: [in a new window] |
TABLE 1. Amino acid sequences of VH CDR3 and usage of VH and VL genes by the isolated scFv against VP8*a
|
VP8* is implicated in the viral recognition of glycoproteins on the surface of mature enterocytes (4), and it is therefore involved in the first steps of viral entry (14, 15). VP8*-specific IgA monoclonal antibodies and mouse lacteal IgG antibodies are able to inhibit in vitro and in vivo rotavirus infections (5, 22), probably by preventing virus attachment to cells. We therefore tested the ability of our scFv to inhibit rotavirus infection in vitro. A representative of each group of clones was chosen (clones 2A1, 2A2, 2A4, 2B3, 2E3, 2E4, 2E11, and 2H6). Phages displaying scFv::pIII fusions (109 PFU) were incubated with trypsin-activated rotavirus strain SA11 (3 x 102 PFU) in 50 µl of PBS, and the mixture was used to infect MA104 cell monolayers grown in microtiter plates in minimal essential medium (MEM) supplemented with 10% fetal bovine serum. The microtiter plates were centrifuged for 1 h at 300 x g, and cells were washed with 100 µl of MEM. Two hundred microliters of MEM containing 1 µg of trypsin/ml without fetal bovine serum was added to each well, and plates were incubated at 37°C for 15 h. Cells were then washed with PBS and fixed for 10 min with 100 µl of acetone-methanol (1:1). Infection spots were developed with goat anti-rotavirus serum (Chemicon) followed by horseradish peroxidase (HRPO)-conjugated anti-goat IgG, and they were counted by using a microscope. It was observed that all of the scFv-carrying phages were able to inhibit viral infection by 40 to 80% compared to an M13 phage displaying an unrelated scFv or a control where no phage was added (Fig. 2). According to Ruggeri et al. (21), a percentage of inhibition of greater than 60% can be considered indicative of a blocking antibody.
![]() View larger version (18K): [in a new window] |
FIG. 2. In vitro inhibition of SA11 virus infection in MA104 cells. Phages carrying scFv::pIII fusions were incubated with the SA11 virus, and the mixture was used to infect MA104 cells. SA11 proliferation was visualized with an anti-rotavirus goat serum and an HRPO-conjugated anti-goat IgG. The results are shown as percentages of infection relative to results of an experiment where no phage was added. The control represents a phage expressing an unrelated scFv.
|
One of the scFv with a high SA11 strain-blocking activity (the 2A1 clone) was chosen, and its gene was cloned in the LAB expression vector pRo266. This vector was obtained by replacing the usp45 secretion and spaX cell wall anchoring signals in pT1NX (23) by the sequence for the signal peptide and cell wall anchor regions from the L. casei cell wall proteinase, PrtP (10). Briefly, the gene fragment from the PrtP cell wall anchor signal (PrtPAnch) was amplified by PCR with a Expand High Fidelity PCR kit (Roche) and with oligonucleotides 5'-CGAGTGGATCCAAGGTACTTGA-3' and 5'-ATGTTACAGCCATCGGTACCGCA-3' and L. casei chromosomal DNA as the template (restriction sites introduced in the oligonucleotides to facilitate cloning are underlined). The PCR product obtained was digested with BamHI and cloned into pT1NX digested with BamHI-SpeI (made blunt with the Klenow enzyme). The resulting plasmid was digested with BamHI and BglII (blunt ended) and ligated to the PCR product encoding the PrtP secretion signal (ssPrtP) obtained with oligonucleotides 5'-GGTTCTAGAACTTTTGGG-3' and 5'-ATGAGGATCCGTCGCCGGCCGAGATAGCCGCCTT-3' and digested with BamHI, resulting in pRo266. The fragment encoding 2A1 scFv was amplified by PCR with oligonucleotides SCFV1 (5'-GCGGCCGGCCCGGCCATGC-3') and Fdseq1 (5'-GAATTTTCTGTATGAGG-3'). It was then digested with NgoMIV and cloned into pRo266 digested with NgoMIV-NcoI (blunt ended), yielding pScFv3, which contained an ssPrtP::scFv fusion. Plasmid pScFv4, carrying an ssPrtP::scFv::PrtPAnch fusion, was constructed by cloning the 2A1 scFv coding region, amplified with oligonucleotides SCFV1 and SCFV3 (5'-CTGCGGCCCCATTCAGATCC-3'), and digested with NgoMIV into pRo266 digested with NgoMIV-BamHI (blunt ended). Ligation mixtures were used directly to transform Lactococcus lactis MG1363, and sequencing of the corresponding plasmids was carried out to verify the correct sequences. Transformation of L. casei with the anchoring vector pScFv4 was very inefficient, and the plasmid showed structural instability in that host (data not shown). L. casei cells harboring pScFv3 were grown in 15 ml of MRS medium for 3 hours and then resuspended in 15 ml of M9 medium supplemented with 0.5% glucose, 0.1% tryptone, 10 mM MgCl2, 1 mM CaCl2, and 5 µg of erythromycin/ml, buffered with 50 mM sodium carbonate (pH 7.4), and grown for a further 4 h. Cell supernatants and crude extracts were prepared as described previously (2). Samples were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and probed by Western blotting with an anti-c-myc monoclonal antibody (1 µg/ml; Roche), which recognizes the C-terminal myc tag present in the scFv (8), followed by incubating with anti-mouse IgG conjugated to alkaline phosphatase and using nitroblue tetrazolium-BCIP (Roche) as the substrate. As can be seen in Fig. 3A, cells were able to secrete mature scFv to the growth medium, although approximately one-half of the protein remained intracellular, as the unprocessed form (Fig. 3A). The scFv secreted by L. casei cells was concentrated 50-fold from the growth medium with an Amicon Ultra centrifugal filter device (10-kDa molecular mass cutoff; Millipore) and used in an ELISA of VP8*. scFv binding was detected with the anti-c-myc monoclonal antibody (4 µg/ml) and a 1:2,000 dilution of alkaline phosphatase-conjugated anti-mouse IgG (Roche). As shown in Fig. 3B, 2A1 scFv was able to recognize SA11 VP8*. This result indicated that the secreted scFv had a correct folding that would retain its biological activity and demonstrated the potential of L. casei to be engineered to express such antibodies. Our future work will be aimed at the optimization of scFv-producing strains and testing them in an in vivo model of rotavirus infection in mice. In particular, new constructs designed to display the scFv on the L. casei surface would be necessary, as cells might function as multibinding reagents in VP8* recognition, increasing their avidity.
![]() View larger version (17K): [in a new window] |
FIG. 3. Expression and functionality of 2A1 scFv in L. casei. (A) Western blot of cell extract and supernatant from L. casei (pScFv3), equivalent to 1 ml of culture, probed with an anti-c-myc tag which recognized the tagged scFv. Molecular weight markers (in thousands) are shown on the left. (B) Functionality of 2A1 scFv secreted by L. casei as determined by ELISA. Culture supernatants of L. casei secreting 2A1 scFv were concentrated 50-fold, and several dilutions were assayed for VP8* recognition by ELISA. The control was supernatants of an L. casei strain carrying the cloning vector pRo266. BSA was used as the negative control.
|
ACKNOWLEDGMENTS
We thank Sue Ellis from the Centre for Protein Engineering (Cambridge, England) for supplying the Griffin.1 library and Lothar Steidler (University of Ghent, Ghent, Belgium) for plasmid pT1NX.
This work was carried out with financial support from the Commission of the European Communities, contract QLRT-2000-0146 (DEPROHEALTH). It was also financed by the Oficina de Ciencia y Tecnología of the Generalitat Valenciana. J. Rodríguez-Díaz was the recipient of a fellowship from the Spanish Ministerio de Educación y Cultura.
This work does not necessarily reflect the views of the Commission of the European Communities and in no way anticipates the Commission's future policy in this area.
REFERENCES
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»