Previous Article | Next Article ![]()
Applied and Environmental Microbiology, January 2006, p. 745-752, Vol. 72, No. 1
0099-2240/06/$08.00+0 doi:10.1128/AEM.72.1.745-752.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Centro de Biotecnologia,1 Laboratório de Genética,2 Laboratório de Biologia Celular, Instituto Butantan, São Paulo, Brazil,3 Interunidades em Biotecnologia, Instituto de Ciências Biomédicas, Universidade de São Paulo, São Paulo, Brazil,4 Instituto Ludwig de Pesquisa sobre o Câncer, São Paulo, Brazil,5 Instituto de Agroquímica y Tecnologia de Alimentos,6 Instituto de Investigaciones Citológicas, Valencia, Spain,7 Instituto de Biociências e Instituto de Química, Universidade de São Paulo, São Paulo, Brazil8
Received 1 June 2005/ Accepted 22 September 2005
|
|
|---|
|
|
|---|
Several immunotherapy approaches to target the development of tumors in infected individuals are based on E6 and E7 oncoprotein expression and presentation by different systems (3, 34, 36). On the other hand, the most promising vaccine for prevention of infection with HPV-16 is based on the HPV-16 L1 major capsid protein (13, 18, 40). L1 spontaneously self-assembles into virus-like particles (VLPs), which are structures that are morphologically similar to native papillomavirus (8, 17). Different approaches for production of HPV-16 VLPs for vaccine purposes have been studied in prokaryotic and eukaryotic expression systems (9, 17, 19, 26, 37, 39, 44). Vaccination with species-specific papillomavirus L1 VLPs protects animals from experimental challenge with infectious virions (e.g., cottontail rabbit papillomavirus in rabbits and canine oral papillomavirus in beagles) (35, 38), indicating that VLPs conserve conformational epitopes present on native virion surfaces that are important for the induction of protective antibodies.
Since the genital mucosa is the host infection site for HPV-16, development of a mucosal vaccine that is capable of inducing a protective HPV-16-specific immune response is a promising strategy. One way to deliver vaccine antigens at the mucosal surfaces is to use live bacterial vaccines. Lactobacilli are gram-positive lactic acid bacteria (LAB) that are classified as generally recognized as safe, a safety status attributed to them because they are commonly used in the food industry and are considered safe organisms for human consumption. In addition, some strains belong to the normal commensal microbial flora of the gastrointestinal and genitourinary tracts of humans. In addition to the safety profiles of Lactobacillus strains, a large variety of probiotic activities and intrinsic adjuvant properties have encouraged studies on the potential of these strains as antigen delivery vectors (22, 31, 33). LAB that express different antigens from human pathogens at different cellular locations have been investigated (3, 28, 41), and the best-characterized studies of the protective efficacy of recombinant LAB involved tetanus toxin fragment C expression (12).
E7 protein was the first HPV-16 antigen expressed in LAB; its expression was described in Lactococcus lactis (3, 4) and Streptococcus gordonii (24). In both cases, specific immune responses were obtained in mice after administration of E7-producing bacteria (4, 24). However, expression in LAB of a potential candidate for a prophylactic vaccine against HPV-16, the L1 protein, has not been reported previously.
In this work, we used a lactose-inducible system based on the Lactobacillus casei lactose operon promoter (11) to express the HPV-16 L1 major capsid protein in L. casei. Here we describe efficient expression of HPV-16 L1 and the ability of this protein to self-assemble into VLPs in L. casei intracellularly. The presence of conformational epitopes in the Lactobacillus-produced VLPs was confirmed by immunofluorescence with an anti-HPV-16 VLP conformational antibody and by an enzyme-linked immunosorbent assay (ELISA) performed with sera from mice subcutaneously immunized with L. casei expressing L1.
|
|
|---|
was grown in LB medium at 37°C with shaking for replication of the expression vectors in the cloning procedures. The erythromycin concentrations used for selection of recombinant strains were 350 µg/ml and 5 µg/ml for E. coli and L. casei, respectively.
HPV-16 L1 amplification, cloning, and generation of recombinant strains.
The HPV-16 L1 gene was amplified by PCR from baculovirus expression plasmid pEVmod/L1 (17), which was kindly provided by Richard Roden (John Hopkins School of Medicine, Baltimore, MD). PCR was performed using high-fidelity Platinum Pfx DNA polymerase (Invitrogen) in a 50-µl reaction mixture containing 1.5 mM MgCl2, 1x Platinum Pfx DNA polymerase buffer, each deoxynucleoside triphosphate at a concentration of 0.3 mM, and 20 pmol of each primer (L1 Forward [5'AGATCTCATATGTCTCTTTGGCTGCCTAGTGAG3'] and L1 Reverse [5'GATATCTTACAGCTTACGTTTTTTG3']). In addition, site-directed mutagenesis was performed to alter two putative transcription termination regions at nucleotide positions 739 to 745 and 1501 to 1506, as described previously (44). For this, two L1 fragments were amplified using primers L1 Forward and MotifT L1 Reverse (5'CATTTGTTCCCTTCGTAAATAGAAG3') and primers MotifT L1 Forward (5'CTTCTATTTACGAAGGGAACAAATG3') and MotifA L1 Reverse (5'GATATCTTACAGCTTACGCTTCTTG3') (for sequences, BglII, NdeI, and EcoRV restriction sites are underlined, and the desired mutations are indicated by boldface type). These two L1 fragments were subsequently used as DNA templates in a PCR with primers L1 Forward and Motif A L1 Reverse to obtain a full-length L1 protein devoid of the internal putative transcription termination sequences. The final PCR product was cleaved with BglII and EcoRV restriction endonucleases and ligated to the pIAlac inducible expression vector (11), which was previously digested with BamHI (coesive ends compatible with BglII ends) and EcoRI (converted to blunt end using the Klenow enzyme), producing the pIAlac/L1 vector.
E. coli competent cells were transformed with ligation mixtures, and erythromycin-resistant clones were screened for the presence of the L1 gene by PCR. The correct sequence was confirmed by DNA sequencing using a 3100 DNA sequencer (Applied Biosystems). The purified pIAlac/L1 plasmid was used for electroporation of L. casei, as previously described (32). Positive E. coli and L. casei clones were frozen in LB and MRS media containing 20% glycerol, respectively, at 80°C.
Expression of HPV-16 L1 in L. casei.
Recombinant erythromycin-resistant L. casei clones carrying the pIAlac/L1 plasmid (L. casei/L1) were grown in basal MRS medium supplemented with 0.5% lactose or glucose at 37°C without shaking until the optical density at 600 nm (OD600) was 2. The L. casei cultures were centrifuged at 5,000 x g for 15 min, and each pellet was suspended (1:20) in phosphate-buffered saline (PBS) containing a protease inhibitor cocktail (Sigma Aldrich). The bacterial suspension was transferred to 2-ml tubes containing the same volume of glass beads (0.5 mm), and a cell extract was prepared by shaking with a Bead-Beater (Biospec Products), using three 30-s cycles at 4,600 rpm.
Total extracts from approximately 109 CFU of induced and noninduced L. casei, as estimated by dilution plating, were separated by gel electrophoresis using sodium dodecyl sulfate-10% polyacrylamide gels and were electrotransferred to nitrocellulose membranes (GE Healthcare). The membranes were blocked overnight at 4°C in PBS containing 0.05% Tween 20 (PBS-T) supplemented with 10% nonfat milk and then incubated for 2 h at room temperature with the HPV-16 L1-specific monoclonal antibody Camvir-1 (Chemicon International) diluted in PBS with 3% bovine serum albumin (BSA). The membranes were washed three times in PBS-T for 10 min and then incubated with peroxidase-conjugated goat anti-mouse immunoglobulin (IgG) (Sigma Aldrich) diluted in PBS-T with 5% nonfat milk for 1 h at room temperature. After washing, the L1 protein was detected using an ECL enhanced chemiluminescence detection kit (GE Healthcare). Quantitative analysis of L1 expression in L. casei was based on a standard curve generated with HPV-16 VLPs expressed in Spodoptera frugiperda cells (kindly provided by R. Roden) and purified by using a previously described protocol (17). A densitometric analysis of the L1 bands observed in recombinant L. casei extracts by Western blotting was performed using software from Eagle Eye still video system (Stratagene). The results described below are the means for L1 expressed in nanograms per 109 CFU from three independent experiments.
Electron microscopy.
L. casei and L. casei/L1 were grown until the OD600 was 2. Cultures (2 ml) were centrifuged at 5,000 x g for 10 min, and each pellet was washed three times in saline. Bacteria were fixed with 3% paraformaldehyde (Sigma) for 1 h at room temperature. Appropriate volumes of fixed intact bacteria were adhered to carbon-coated grids for 1 min and air dried. Some samples were negatively stained with 2% uranyl acetate for 1 min. The samples were examined with a Zeiss EM 109 transmission electron microscope operated at 80 kV.
Immunofluorescence.
L. casei and L. casei/L1 were washed and fixed as described above. Subsequently, the cells were adhered to polylysine-coated microscopic glass slides. To detect L. casei-produced VLPs by immunofluorescence, the slides were incubated with an anti-HPV-16 VLP conformational antibody, H16.V5 (10) (kindly provided by N. Christensen, Department of Microbiology and Immunology, Pennsylvania State University College of Medicine, Hershey, PA), diluted in PBS containing 0.01% Tween 20 and 0.5% BSA for 2 h at 37°C. The cells were then washed five times in PBS for 5 min and incubated with fluorescein isothiocyanate (FITC)-conjugated anti-mouse IgG (Sigma) diluted in PBS containing 0.01% Tween 20 and 1.5% BSA for 1 h at 37°C. After five washes in PBS, the slides were air dried, embedded in Mowiol (Calbiochem), covered with coverslips, and stored at 4°C until use. The fluorescence was visualized by exposure of the slides to a filter with an excitation wavelength of 488 nm from a Zeiss LSM 510 META confocal microscope.
Sucrose gradient sedimentation.
One-half liter of an induced-L. casei/L1 culture was centrifuged at 5,000 x g for 20 min, and the bacterial pellet was suspended in 10 ml of PBS-0.5 M NaCl buffer containing a protease inhibitor cocktail (Sigma). The bacterial suspension was disrupted with a French press (Thermo Electron Corporation), using two cycles at 2,000 lb/in2. An aliquot of cellular extract (approximately 1011 CFU) was applied slowly to the top of a 10 to 65% sucrose gradient (12 ml) prepared in PBS-0.5 M NaCl buffer. The gradient was centrifuged at 154,000 x g in an SW40Ti rotor (Beckman) at 4°C for 1.5 h and then fractionated into 23 aliquots (approximately 500 µl each) from the bottom to the top of the centrifuge tube (the sucrose concentrations of all fractions were calculated using a refractometer). The odd fractions were then analyzed by Western blotting with the anti-HPV-16 L1 Camvir antibody. The 30 to 40% sucrose fractions containing L1 were pooled, dialyzed against PBS-0.5 M NaCl, and centrifuged at 183,000 x g at 4°C for 1.5 h, using a 70Ti rotor (Beckman). The pellet was suspended in 100 µl of PBS-0.5 M NaCl, and an aliquot was adsorbed on a carbon-coated grid, air dried, and negatively stained with 2% phosphotungstic acid (pH 7.0) for 1 min. The adsorbed protein was analyzed with a LEO 906E transmission electron microscope.
Immunization of mice.
Six- to eight-week-old female BALB/c mice from Instituto Butantan were used for subcutaneous immunization experiments. L. casei and L. casei/L1 were grown until the OD600 was 2. Bacteria were collected by centrifugation at 5,000 x g and washed three times with apyrogenic saline, and then the concentration was adjusted to 109 cells per 100 µl. Groups of five mice were inoculated with 100 µl of saline, L. casei, or L. casei/L1 on two consecutive days. Three administrations, one priming and two boosts, were performed at 2-week intervals, for a total of six administrations. Ten days after the last administration, animals were bled through the retroorbital plexus, and pooled sera were collected and stored at 20°C until use. The results described below are representative of two experiments.
ELISA for detecting anti-HPV-16 VLP antibodies.
ELISA plates were coated with 100 ng of S. frugiperda insect cell-derived VLPs diluted in PBS (assembled VLPs) or in 0.2 M carbonate buffer (pH 9.6) (disassembled VLPs) (23) at 4°C for 16 h. Pooled sera were tested for the presence of anti-L1 IgG as previously described (1). The titer was defined as the dilution at which the absorbance at 492 nm was 0.15. The results presented below are the means of two independent immunization experiments.
|
|
|---|
Intracellular expression of the HPV-16 L1 protein was tested by Western blotting using L. casei/L1 extracts. A 56-kDa protein was specifically recognized by the anti-HPV-16 L1 monoclonal antibody in extracts from a representative clone grown in basal MRS medium containing lactose (Fig. 1, lane 2). L1 immunoreactive protein bands at lower molecular masses were often observed, as described previously for other expression systems (17, 44), which may have resulted from degradation of full-length L1 protein. The predominant 45-kDa degradation product coincided with the previously well-characterized fragment derived after digestion of the whole L1 protein with trypsin (7, 21), indicating possible action of L. casei serine proteases.
![]() View larger version (53K): [in a new window] |
FIG. 1. Lactose-inducible expression of HPV-16 L1 protein in L. casei/L1. Total extracts from L. casei/L1 clones grown in the presence of lactose (lane 2) or glucose (lane 1) were analyzed by Western blotting using the anti-HPV-16 L1 monoclonal antibody Camvir. The L1 56-kDa band is indicated by an arrow. L. casei (lane 3) and recombinant L. casei expressing an unrelated protein (S. pneumoniae PsaA protein) (lane 4) grown in lactose-containing medium were used as negative controls.
|
Intracellular production of HPV-16 L1 virus-like particles by L. casei.
After characterization of HPV-16 L1 expression in L. casei, we investigated whether L. casei was able to produce L1 VLPs intracellularly. After growth under induction conditions, L. casei and L casei/L1 cells were prepared for intact bacterial ultrastructural analysis, as described in Materials and Methods.
L. casei/L1 cells exhibited different VLP production patterns. While a few cells exhibited large amounts of VLPs and electron-dense bodies containing VLP aggregates (Fig. 2A and C), other cells exhibited reduced production. By scanning the grids, it was also possible to observe disrupted L. casei/L1 cells containing dispersed VLPs, which facilitated observation of the particles (Fig. 2D). The sizes of the L. casei-produced VLPs were heterogeneous; these VLPs ranged from 30 to 60 nm in diameter, in accordance with the 55-nm particle pattern obtained in eukaryotic L1 expression systems (17). No VLP-like structures were observed in wild-type L. casei (Fig. 2B).
![]() View larger version (66K): [in a new window] |
FIG. 2. Intracellular production of HPV-16 L1 VLPs by L. casei/L1. (A and B) Electron micrographs (magnification, x108,000) of L. casei/L1 (A) and wild-type L. casei (B) grown in basal MRS medium containing lactose, fixed with 3% paraformaldehyde, and negatively stained with 2% uranyl acetate. Dispersed VLPs (arrows), as well as VLPs localized in electron-dense bodies (arrowheads), are present only in L. casei/L1. (C) For better visualization of L. casei/L1 cytoplasm, fixed bacteria were not negatively stained with uranyl acetate. The micrograph (magnification, x88,500) shows part of an L. casei/L1 cell with clear examples of electron-dense bodies containing VLP aggregates (arrowheads). CM, cytoplasmic membrane. (D) Scanning electron micrograph of a grid region with disrupted L. casei/L1 containing VLPs (magnification, x141,000). The VLPs analyzed had diameters ranging from approximately 30 to 60 nm. (A, B and C) Bars = 250 nm; (D) bar = 100 nm.
|
![]() View larger version (23K): [in a new window] |
FIG. 3. L. casei-produced VLPs display the H16.V5 epitope present on HPV-16. (B and D) Immunofluorescence microscopy of L. casei/L1 (B) and wild-type L. casei (D) cells harvested after lactose induction and fixed with 3% paraformaldehyde. Both types of cells were treated with the H16.V5 conformational antibody and FITC-conjugated goat anti-mouse IgG. Only L. casei/L1 displayed bright immunostaining in specific sites (arrowheads), possibly associated with electron-dense bodies. Panels B and D are views of panels A and C, respectively, after exposure to an appropriate filter for FITC. Bars = 3 µm.
|
Characterization of L. casei-produced VLPs by sucrose gradient sedimentation.
The sedimentation profile of L. casei/L1 extracts in a 10 to 65% sucrose gradient was also analyzed to detect the presence of VLP-structured L1. Odd fractions, collected after centrifugation, were analyzed by Western blotting using the anti-L1 Camvir antibody (Fig. 4A). L1 produced bands from the middle sucrose gradient fractions to the top. This result suggests that the L. casei-produced L1 protein consisted of high-density molecules typical of VLPs (fractions 14 to 17), which sedimented in the 30 to 40% sucrose fractions, as previously described (23), as well as in disassembled intermediary structures (fractions 18 to 22) and L1 monomers (fraction 23). Higher-density fractions (fractions 14 to 17) (Fig. 4A) containing L1 were pooled, and analysis of this pool by electron microscopy indicated the presence of VLP-like ordered particles (Fig. 4B).
![]() View larger version (36K): [in a new window] |
FIG. 4. Fractionation of L. casei/L1 extract in a 10 to 65% sucrose gradient. (A) Fractions were obtained from the bottom to the top of the gradient. Lane 1 contained the heaviest fraction. The L1 protein was detected by Western blotting with the anti-HPV-16 L1 Camvir antibody (arrow). The negative and positive controls were wild-type L. casei extract (lane 13) and L. casei/L1 extract (lane 14), respectively. (B) High-density fractions (fractions 14 to 17) containing L1, corresponding to 30 to 40% sucrose, were pooled and analyzed by electron microscopy. The sizes of L. casei-produced VLPs (arrows) were heterogeneous, ranging from 20 to 60 nm. Bar = 100 nm.
|
![]() View larger version (15K): [in a new window] |
FIG. 5. Production of conformational and nonconformational antibodies in mice subcutaneously immunized with live L. casei/L1. ELISA plates were coated with 100 ng of S. frugiperda-produced VLPs diluted in PBS (assembled VLPs) (A) or in 0.2 M carbonate buffer (pH 9.6) (disassembled VLPs) (B), and sera from mice subcutaneously immunized with L. casei/L1, L. casei, or saline were tested using twofold serial dilutions from 1:10. The data are the means of two independent immunization experiments.
|
|
|
|---|
The L1 protein has been expressed previously in other prokaryotic expression systems, such as E. coli (44) and Salmonella enterica serovar Typhimurium (26). In the former organism, the L1 protein was expressed in insoluble inclusion bodies, and in vitro refolding protocols were necessary to produce VLPs. In the latter organism, intracellular VLP production was identified by using a sucrose gradient with recombinant bacterial extracts, although VLPs were not structurally characterized in intact bacteria.
In this work, using an inducible expression system, we expressed the HPV-16 L1 protein in L. casei and obtained different kinds of evidence for intracellular formation of L1 VLPs. The production of VLPs by L. casei is an interesting result, since assembly of L1 into VLPs requires disulfide bond formation (15), which is a posttranslational modification generally not observed in the bacterial reductive cytoplasmic environment. Nevertheless, cytoplasmic production of disulfide bond-dependent proteins in prokaryotic cells has been reported previously (27). Very little is known about the catalysis of disulfide bond formation in gram-positive bacteria, and most of the data have been obtained with Bacillus subtilis. Analyzing the partial genome data of L. casei ATCC 334 (http://genome.jgi-psf.org/draft_microbes/lacca/lacca.home.html), we found that some putative disulfide oxidoreductases and disulfide isomerases were annotated, but none of these enzymes has been functionally characterized yet. Some of them exhibited homology to B. subtilis BdbA (5) and to Bacillus brevis Bdb (14), which was the first disulfide oxidoreductase characterized in gram-positive bacteria and has a thioredoxin-like fold and a CXXC motif as the active site. No homology was observed with other B. subtilis Bdb family proteins (5) or with E. coli Dsb family proteins (2). Besides the possibility of cytoplasmic disulfide oxidoreductase action in Lactobacillus, the L1 protein was localized at preferential sites, characterized as electron-dense bodies, which could help capsid assembly in the cytoplasmic environment by formation of a "subcellular environment." Disulfide bond-independent capsid assembly in bacterial cytoplasm has been described previously (25). Moreover, production of HPV-16 L1 VLPs in Salmonella has been described, and the VLPs were ultrastructurally characterized after sucrose density gradient sedimentation (26).
Cells with large amounts of VLPs and electron-dense bodies were not frequently found, but we observed that most of these cells were larger than normal cells (Fig. 2C, 3A, and 3B). Bacterial filamentation has been described as a defect in completing replication and cell division and has been observed previously in bacteria responding to a variety of stresses, such as the production of large amounts of heterologous proteins, commonly described in E. coli (16). In fact, it took recombinant L. casei/L1 clones 12 h longer than it took the wild-type strain to reach an optical density of 2.0 (data not shown). Studies with Lactobacillus strains revealed cell filamentation in a Lactobacillus plantarum cell wall mutant (29) and in Lactobacillus alimentarius subjected to stress after treatment with organic acids (20). However, to our knowledge there are no published data about cell elongation during the production of recombinant proteins in Lactobacillus.
HPV-16 L1 VLPs that are produced in eukaryotic cells, such as Saccharomyces cerevisiae cells (39) and insect cells (17), display conformational epitopes present on HPV-16 viral particles, which are important for eliciting virus-neutralizing antibodies. To test whether such conformational epitopes were displayed on Lactobacillus-expressed VLPs, an immunofluorescence assay was performed using the anti-HPV-16 VLP conformational antibody H16.V5 (Fig. 3). This antibody recognizes the H16.V5 epitope of HPV-16 and has been characterized as the HPV-16 major neutralizing antibody, which competes with about 75% of all patient antisera (42). Moreover, analysis of HPV-16 L1 mutants showed that, besides the H16.V5 epitope, few additional potent neutralization sites were presented on HPV-16 VLPs (43). The immunofluorescence data showed that H16.V5 antibody recognized ordered particles produced by L. casei/L1, which were predominantly associated with electron-dense bodies (Fig. 3B).
Anti-L1 IgG was detected in sera obtained from mice subcutaneously immunized with L. casei/L1. These sera reacted with insect cell-produced VLPs, suggesting that the L. casei-expressed L1 protein adopted a native conformation. This result, together with the electron microscopy, immunofluorescence, and sucrose gradient sedimentation data, indicates that L. casei-expressed L1 protein may have formed VLPs resembling authentic HPV-16 L1 VLPs.
Nasal inoculation of mice with L. casei/L1, using the protocol described for subcutaneous immunization, did not produce any mucosal or systemic humoral response against L1 (data not shown). There could be many reasons for this lack of immune response, including (i) insufficient L1 protein presented by recombinant lactobacilli (approximately 40 ng/109 CFU L. casei/L1) at mucosal sites to elicit a consistent immune response; (ii) the absence of the inductor (lactose) by the time of inoculation in the animal; and (iii) the strain used, since the lac-inducible expression system is specific for L. casei, limiting the use of other LAB strains. Thus, other approaches to improve mucosal immunization with recombinant LAB expressing L1, such as the use of a constitutive expression system, need to be examined further. Nevertheless, the possibility of toxicity of L1 VLPs in a constitutive system, as well as production of even lower levels of the recombinant protein, cannot be ruled out. Based on data for the use of Lactobacillus as a live vaccine vector, the minimum amount of antigen that could induce immune responses through mucosal presentation is not clear, and it is probably dependent on the antigen and the strain used (22).
In fact, the results obtained with this inducible system are, as far as we know, the first results for the potential of LAB to express L1 and to produce VLPs. The development of a mucosal vaccine against HPV using Lactobacillus expressing L1 assembled into VLPs should be investigated further.
This work was supported by FAPESP, CNPq, and Fundação Butantan.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»