ETEC strains were isolated from children with diarrhea. The strains were previously characterized for the presence of CFs and enteroxins and used for challenge purpose. Plasmids pET-28a and pET-32a were from Novagen (USA). E. coli BL21 (DE3), DH5α and Top10 were procured from Pasteur Institute of Iran. The E. coli strains were grown in Luria–Bertani (LB) broth at 37°C.

Chimeric gene was designed based on the nucleotide sequence of ltb, cssA and cssB genes (Gen Bank accession numbers; M17874.1, GQ241328.1, UO4844.1, respectively). The sequences were used to generate a chimeric trivalent protein fused together by (EAAAK)4 hydrophobic linkers in order to find the best epitope exposing chimeric antigen. The in silico gene analysis and multi parameter gene optimization of the synthetic chimeric gene was performed using OPTIMIZER server. Itasser and phyre software were used in order to estimate tertiary structure prediction. The chimeric proteins were analyzed for B-cell epitopes using Bcepred server. The gene encoding protein was synthesized by Shine Gene Molecular Biotech, Inc. (Shanghai, China).

The synthetic gene (cssA- cssB -ltB) subcloned into pET28a was transformed to E. coli BL21 (DE3) (Novagen). The transformants were grown in Luria Bertani (LB) broth supplemented with 30μg of kanamycin/ml. Expression of the chimeric protein was induced at OD600 of 0.5 by addition of 1mM isopropyl-β-d-galactopyranoside (IPTG) and analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Batch purification of His-tagged proteins from E. coli under denaturing conditions (Qiagen) was carried out and validated by SDS-PAGE. The denaturant (8M urea) was removed by stepwise dialysis.

The proteins separated by SDS-PAGE were blotted onto nitrocellulose membrane using transfer buffer (39mM glycine, 48mM Tris–base, 0.037% SDS, and 20% methanol). The membrane was incubated in 10ml blocking buffer (5% skim milk in PBST [137mM NaCl, 2.7mM KCl, 4.3mM Na2HPO4 and 0.05% (v/v) Tween-20]) at 4° C overnight. The membrane was then incubated in a 1:10,000 dilution of mice anti-His-tag -conjugated HRP and anti-CTX (Sigma) in the PBST, with mild shaking for 1h at 37°C. Detection was carried out using HRP staining solution containing 3,3′-diaminobenzidine (DAB) (Sigma) and H2O2.

Five-week old female BALB/C mice and New Zealand White female rabbits procured from Pasteur Institute of Iran were randomly divided into test and control groups. Animals were immunized four times. Each mouse received 20μg recombinant CS6 protein along with complete Freund's adjuvant (SIGMA) subcutaneously. Booster doses of 15 and 10μg recombinant proteins with incomplete Freund's adjuvant were injected after 20 and 35 days, respectively. PBS served as control in parallel to the aforementioned procedure. The sera taken from animal blood after the second and third injections were stored at −70°C for further analyses.

Indirect ELISA was used for antibody titration. A 96-well ELISA plates (Nunc) were coated with 5μg of CssA, CssB and LTB proteins in coating buffer (64mM Na2CO3, 136mM NaHCO3, pH 9.8) at 4°C overnight. The plates were washed three times with PBST and the non-specific sites were blocked with skim milk solution 5% (w/v) in PBST. The serum samples serving as primary antibodies were serially diluted to 1:500 in PBST and added to the ELISA plates with incubation at 37°C for 45min. The plates were washed thrice in PBST and Goat Anti-Mouse IgG Antibody (12,000 in PBST) was added to the ELISA plates as the secondary antibody. The plates were incubated for 30min at 37°C and were then washed. The wells added with 100μl of citrate buffer containing 0.06% (W/V) of O-phenylene diamine dihydrochloride (OPD) (SIGMA) and 0.06% (V/V) hydrogen peroxide were incubated at room temperature for 15min. The reaction was stopped with 100μl of 2M H2SO4 and the OD492 was read on a microplate reader (Bio-Rad).

Caco-2 cells (104/well) were added and grown to confluence in 96-well plates with DMEM supplemented with 10% (v/v) FBS at 37°C in a culture flask. ETEC strain was grown in LB broth. The bacterial suspension (105/well) was pretreated with 40μl of immunized mice antisera for 15min at room temperature. The bacterial mixture was added to Caco-2 cells and incubated for 2h at room temperature. The supernatant was decanted, and the cells were washed with PBS. After digestion, the suspended bacteria were serially diluted and incubated in LB agar at 37°C for 16h. Subsequently, the number of bacteria adhered to the Caco-2 cells was counted. The sera of non-immunized mice were used as control.

The production of LT toxin neutralizing antibodies was performed with rabbit ileal loop, cAMP immunoassay and Y-1 cell rounding assay. The rabbit ileal loop test was carried out according to the methods described by De and Chatterjee.18 New Zealand White rabbits (1.5kg) were fasted for 24h before surgery. The rabbits were anesthetized by the intramuscular injection of fentanyl (SIGMA) (0.3ml/kg of body weight) and zolazepam chloridrate (0.4ml/kg of body weight). 10-cm ligated ileal loops were constructed. ETEC at 1×108 concentration were incubated with mouse serum raised against the chimeric protein, for 30min. ETEC bacteria incubated with non-immune mouse serum served as negative control. The above bacterial mixture and PBS were inoculated into each ligated loop using a 25-gauge needle. After 18h, the animals were sacrificed with intravenous 3% pentobarbital, and zolazepam chloridrate (0.4ml/kg). For each ileal loop, the volume of secretion was measured after excision. Experimental infection of the rabbits was performed at the Faculty of Veterinary Medicine, University of Tehran. Research was conducted in compliance with the Animal Welfare Act and regulations related to experiments involving animals. The production of antibodies that neutralized LT toxin was also performed with a cAMP immunoassay kit (Abnova) and CHO cells. CHO cells were seeded in 24-well sterile culture plates at a density of 5×104 cells per well and were grown overnight to approximately 80% confluence. 0.1μg of LT was incubated with 50μl mice serum at room temperature for 1h, and the mixture was added to each well for incubation at 37°C in 5% CO2 for 2h After washes, the cells were lysed with 0.1M HCl and the supernatant was used in cAMP assay following the manufacturer's protocol.

For cell rounding assay, Y-1 mouse adrenal cells (Pasteur Institute of Iran) were seeded in cell culture plates at a concentration of 5×104 cells per well and grown to confluence at 37°C/5% CO2 atmosphere. Y-1 cells were mixed with LT pretreated with serum samples and incubated under 5% CO2 incubator at 37°C for 10h. The morphological changes were observed under the compound microscope.

All statistical analyses were conducted using a SPSS 12.0 statistical program. Student's t-test was used to analyze the data for antibody responses between immunized and non-immunized groups; t-test was also used to evaluate the significance of differences in inhibition of ETEC binding to Caco-2 cells generated by test sera. A value of p<0.05 was considered statistically significant.

A chimeric trivalent protein fused together by (EAAAK)4 hydrophobic linkers in order to find the best epitope exposing chimeric antigen (Fig. 1A and B). Codon optimization was performed to improve the transcription efficiency and transcript stability. Percentage of codon having a frequency distribution of 91–100 in the native chimeric CS6 gene was 66%, which was significantly improved to 91% in the optimized gene sequence. The overall GC content, which is a measure of transcriptional and translational efficiency, was improved from 38.96% to 48.75% upon codon optimization. There were five negative cis elements in the native CS6 gene sequence, which were removed after optimization.

Fig. 1.

Design and modeled tertiary structure of chimeric protein: (A) schematic construction of recombinant chimeric protein containing CssB, CssA and LTB bound together by appropriate linkers, (B) amino acid sequences of chimeric protein and (C) modeled tertiary structure of chimeric protein by I-TASSER software. This model showed a protein with three main domains linked together with linkers.

(0.34MB).

Ab initio modeling of the synthetic sequence was exploited to produce 3D models of the chimeric protein. The result of tertiary structure of the chimeric protein construction using I-TASSER showed a protein with three main domains linked together with linker (Fig. 1C). B-cell epitope was predicted and the epitopes having cutoff values >0.8 were selected (Table 1).

The genes were subcloned into pET28a and pET32a vectors respectively and over expressed in E. coli (BL21DE3) (Fig. 2A). The SDS-PAGE analysis showed the presence of a 47 KD recombinant chimeric protein. The purification of recombinant proteins was carried out under denaturation condition by Ni-NTA affinity (Fig. 2B). The expression of the recombinant protein was confirmed by the detection of the protein by Western blot using anti-His-tag and anti CTX antibodies (Fig. 3).

Fig. 2.

Expression and purification of recombinant chimeric protein: (A) SDS-PAGE analysis of the rCS6 expression after IPTG induction. Lane 1: protein weight marker. Lanes 2, 4: non induced cells. Lane 3: the induced bacteria. (B) Purification of rCS6 protein by Ni-NTA column. Lane 1: protein weight marker. Lane 2: expression of chimeric protein. Lane 3: flow-through. Lane 4: column washed with buffer D (pH=5.9). Lane 5: column washed with elution buffer (pH=4.5). Lane 6: column washed with mes Buffer.

(0.15MB).

Fig. 3.

Western blot analysis rCS6: (A) using anti 6X-His-tag antibodies. Lane 1: protein weight marker. Lane 2: rCS6 and (B) using anti-CTXB. Lane 1: protein weight marker, Lane 2: rCS6, Lane 3: antibody. Negative control (bacterial lysate without IPTG induction).

(0.04MB).

Animals were immunized with the recombinant chimeric protein. All immunized animals had anti-CssA, anti-CssB, and anti-LT IgG antibodies detected in serum (Fig. 4).

Fig. 4.

Serum anti-LTB, -CssB, and -CssA antibody titers of mice immunized with chimeric protein. Titers calculated as log 10 of last reciprocal dilution above cut-off. Error bars represent mean±SEM (n=3).

(0.06MB).

Assessment of Caco-2 cell monolayers revealed that ETEC cells pretreated with serum from non-immunized mice were distributed on the Caco-2 cells. However, pretreatment of the ETEC cells with immunized mice antisera considerably blocked their binding to Caco-2 cells. The immune sera could inhibit 68±3.5% bacterial adhesion onto the target cells. Comparison of pretreatment of the bacteria with non-immune antisera with that of immunized mice serum, showed a considerable reduction in their adhesion to Caco-2cells followed by pretreatment of the ETEC cells with immunized mice antisera.

For each ileal loop, the volume of secretion measured after excision was as follows: ETEC as a positive control: 2.59±0.3g/cm, ETEC+ antisera: 0.51±0.01g/cm, PBS: 0.31±0.12g/cm. The fluid accumulation was not observed in rabbit ileal loops 18h post-infection with ETEC treated with the immunized mice serum. In contrast, infection with standard ETEC was accompanied by an increase of fluid secretion. Data from the cAMP immunoassay kit showed that antibodies in serum and fecal samples of the immunized mice neutralized LT and prevented the toxin from stimulating intracellular cyclic AMP levels in CHO cells (Fig. 5). No morphological changes were observed in Y1 cells incubated with LT pretreated with anti-chimeric antibody, whereas normal serum had no inhibitory effects on rounding (Fig. 6).

Fig. 5.

Anti-LT antibodies neutralization assay with cAMP immunoassay. Serum samples from immunized or from the control mice were tested to neutralize LT toxin. Cyclic AMP concentrations of treated CHO cells were measured using a cAMP kit. Mean values are shown, and error bars represent standard deviations.

(0.05MB).

Fig. 6.

Inhibitory effect of anti-L2C3 antibody on the rounding of Y-1 cells. Y-1 cells were co-cultured with LT pretreated with serum of immunized mice (A) or normal serum (B).

(0.26MB).