Investigation into immunological responses against a native recombinant CTB whole-cell Vibrio cholerae vaccine in a rabbit model



Bita Bakhshi, Department of Bacteriology, Faculty of Medical Sciences, Tarbiat Modares University, Jalal-Ale-Ahmad Ave., Tehran 14117-13116, Iran. E-mail:



The aim of this study was to express and purify the recombinant CTB (rCTB) protein from Vibrio cholerae and investigate the biological and immunological characteristics of purified protein in rabbit animal model and in combination with Iranian inactivated V. cholerae whole cells as a domestic recombinant WC-CTB vaccine.

Methods and Results

Expressed 6XHis-tagged rCTB was properly purified, and its identity was confirmed by Western blotting using cholera toxin-specific antibody. Concentration of purified protein was assessed to be 700 mg l−1. GM1-ELISA assay showed that purified rCTB pentamer was functionally active and able to bind GM1 in a dose-dependent manner. Recombinant CTB was inoculated into rabbits through intestinal rout alone and in combination with inactivated whole-cell V. cholerae strains (WC). The anti-CTB IgG titre showed that serum IgG responses were significantly increased in groups immunized with rCTB mixed with inactivated WC in comparison with control group. Furthermore, rCTB without V. cholerae WC also stimulated the IgG responses when inoculated into rabbit intestine. Challenge experiments of immunized rabbits showed an adequate protection against V. cholerae strains.


Recombinant CTB alone and in combination with inactivated Iranian strains was protective against live toxigenic V. cholerae strains, made it a potential candidate for an indigenous vaccine.

Significance and Impact of the Study

It was proved that rCTB produced in this system can be used as a potent immunogenic protein to stimulate the immunity against V. cholerae strains and can be used for developing a native vaccine composed of our local strains with their own surface structures and antigenic determinants against cholera.


Cholera is a severe diarrhoeal disease and an important cause of morbidity and mortality in many Asian countries including Iran (Sanchez and Holmgren 1986; Bakhshi et al. 2008b; Bakhshi and Pourshafie 2009). Cholera can spread rapidly and remains a global threat to countries with poor water sanitations (Yan et al. 2007)

The main virulence element of pathogenic Vibrio cholerae is cholera toxin (CT) that is responsible for cholera symptoms. CT is composed of two different subunits. The subunit A (CTA) is a single copy protein with an enzymatic toxic ability that activates cellular adenylate cyclase level in intestinal cells and accelerates the secretion of chloride and bicarbonate from the mucosal cells to the intestinal lumen and eventually causes diarrhoea. B subunit (CTB) is an 11·6-kDa pentameric protein responsible for binding to GM1 ganglioside receptor on the surface of host intestinal epithelial cells and facilitates endocytosis of CTA to these cells. CTB is recognized as a potent immunogen in the intestinal and nasal mucosal sites (Waldor and Mekalanos 1994). It also can induce oral tolerance when conjugated to antigens, so it can be used as a transmucosal carrier delivery system (Sun et al. 1994; (Czerkinsky et al. 1996); Sun et al. 1996). CTB has been reported as a great adjuvant for vaccines because it can enhance the immune response and elicit serum and secretory antibodies against antigens that are usually poor immunogens (Tuchikubo and Yasuda 2000). The studies indicated that this protein can stimulate the immune system significantly and create immunity against V. cholerae.

Development of an effective vaccine is one of the ways to prevent or control cholera. Many cholera vaccines including killed whole-cell vaccines single handedly or in combination with various adjuvants or purified lipopolysaccharide vaccine have been developed (Liang et al. 2003). Field trials showed that a vaccine consisting of whole cells in combination with B subunit produced limited protection in children younger than 5 years (Taylor et al. 1994; Ryan and Calderwood 2000). Produced vaccines against cholera have been developed over time, and with the advent of recombinant DNA knowledge, different types of potent recombinant strains of V. cholerae were emerged to use for producing vaccines (Levine and Kaper 1996). For achieving proper immune response, proper antigens should also be delivered as killed whole-cell suspension (Thungapathra et al. 1999), which is composed of admix of strains isolated from the region that vaccine is to be used.

The aim of this study was to express and purify the recombinant CTB (rCTB) protein from V. cholerae and investigate the biological and immunological characteristics of purified protein in rabbit animal model and in combination with Iranian inactivated V. cholerae whole cells as a domestic recombinant whole-cell CTB (WC-CTB) vaccine.

Materials and methods

Bacterial strains and plasmids

Bacterial strains and plasmids used in this study are listed in Table 1. Escherichia coli BL21 (DE3) strain was used as an expression host for rCTB production. Killed V. cholerae strains were used for incorporation into WC-CTB admix to immunize rabbits. Vibrio cholerae live strains were used to challenge the immunized rabbits. The pAE plasmid and pAE-ctxB construct were kindly provided by Dr Arêas from Butantan Institute, Brazil.

Table 1. Bacterial strains and plasmids used in this study
  1. a

    Ampr, ampicillin resistant.

Vibrio Cholerae
Clinical isolate from Iran-44Biotype El Tor, Serotype Ogawa O1
ATCC14035Biotype Classic, Serotype Ogawa O1
Clinical isolate from Iran-65Biotype El Tor, Serotype Inaba O1
Escherichia coli
BL21 (DE3) 
pAEAmpra, E. coli plasmid
pAE-ctxBAmpr, E. coli plasmid containing ctxB gene

Production of rCTB in large scale

BL21 (DE3) competent cells were used as recipient of pAE-ctxB construct plasmid. Two hundred millilitres of LB broth containing 100 μg ml−1 ampicillin was inoculated with 5 ml of fresh cultures of BL21 (DE3) cells containing pAE-ctxB plasmid. The cultures were shaken at 180 rev min−1 in 37°C, and isopropyl-β-d-1 thiogalactopyranoside (IPTG) was added to the medium at final concentration of 1 mmol l−1 when the A600 reached 0·8. After cell harvest, by centrifugation, pellets were lysed by 20 ml lysis buffer pH 8·0 (300 mmol l−1 NaCl, 100 mmol l−1 Tris–Cl and 1% v/w Triton X-100) and subjected to sonication for 10 min. After centrifugation, inclusion bodies were washed with binding buffer I (500 mmol l−1 NaCl, 50 mmol l−1 Tris–Cl, 10 mmol l−1 β-mercaptoethanol and 2 mol l−1 urea) and centrifuged at 8422 g for 15 min in each time. Inclusion bodies were dissolved with binding buffer II with 8 mol l−1 urea. The solubilized pellet was slowly diluted in 400 ml of binding buffer III containing 5 mmol l−1 imidazole for the refolding of rCTB and incubated at room temperature for 24 h (Arêas et al. 2002).

Purification of recombinant CTB

In pAE_ctxB construct, rCTB was expressed with 6XHis-tag at N-terminus to facilitate the purification of recombinant protein through Ni2+-charged column chromatography (Arêas et al. 2002). Resin (Sigma-Aldrich, Schnelldorf, Germany) was packed into column and was calibrated. Diluted solubilized suspension was added to column, and after the adsorption of protein, resin was washed with 20 ml of washing buffer (50 mmol l−1 Tris and 10 mmol l−1 imidazole) to wash the nonspecific bindings from resin. Recombinant CTB was eluted with elution buffer (300 mmol l−1 NaCl and 50 mmol l−1 Tris–Cl) containing 1 mol l−1 imidazole.

SDS-PAGE and Western blotting

Boiled protein fractions were mixed by sample buffer and electrophoresed on two identical 15% (v/w) SDS-PAGE gels on a double-sided tank. One gel was stained with 1% (v/w) Coomassie Brilliant Blue R-250 for 2 h, and the other was used for blotting the proteins onto polyvinylidene difluoride (PVDF) using semidry blotting system (Bio-Rad, Hercules, CA, USA). PVDF membranes were blocked in 1% (v/w) nonfat milk powder and incubated in 1 : 1000 dilution of rabbit polyclonal anti-CT antibodies in 0·5% (v/w) Tween 20/PBS (PBS-T) for 2 h at room temperature. Membranes were washed three times with 0·05% (v/w) PBS-T and incubated in 1 : 10 000 dilution of HRP-conjugated goat anti-rabbit IgG (Sigma-Aldrich) in 0·05% (v/w) PBS-T for 30 min. After washing steps, production of rCTB was detected by chemiluminescent substrate (ECL). Chemiluminescence on PVDF membrane was detected by exposure to X-ray films (Kodak, Rochester, NY, USA).

Ganglioside GM1–enzyme-linked immunosorbent assay (GM1-ELISA)

The ability of rCTB to bind to its cellular receptor (GM1) was determined by GM1-ELISA assay. Briefly, 12 wells of 96-well microplate were coated with 1 μg per well of GM1 (Sigma-Aldrich) in carbonate–bicarbonate buffer pH 9·6 and incubated overnight at 4°C. Plates were washed three times with PBS-T (PBS + 0·05% v/w Tween 20) and were coated with different antigens including commercial CTB (cCTB) (Sigma-Aldrich) as positive control and rCTB at final concentration of 1 μg per well and incubated overnight at 4°C. PBS was used as negative control, and each assay was performed in triplicates. After three washing steps in PBS-T, wells were blocked with 300 μl of blocking buffer (1% v/w BSA in PBS) for 2 h at 37°C. Wells were washed and incubated with 100 μl of 1 : 1000 dilution of a polyclonal rabbit anti-CT antibody per well for 3 h at 37°C, followed by three washing steps with PBS-T. HRP-conjugated goat anti-rabbit IgG was added (100 μl per well of 1 : 10 000 dilution) and incubated for 1 h at 37°C. HRP-conjugated antibodies were visualized by the addition of 100 μl per well of o-phenylenediamine (4 mg OPD in 10 ml of 0·2 mol l−1 citrate–phosphate buffer, pH 5·0, in the presence of 4 μl of fresh H2O2 immediately before use). After 15-min incubation at room temperature, the reaction was stopped by adding 50 μl per well 2 mol l−1 H2SO4, and then, the absorbance was measured at 450 nm by ELISA reader.

Antigen preparation for animal assay

Recombinant CTB, cCTB and whole-cell-inactivated toxigenic V. cholerae strains (V. cholerae O1 serotype Ogawa biotype classic ATCC 14035, V. cholerae O1 Ogawa El Tor and V. cholerae O1 Inaba El Tor) were used in this assay (Table 1). Four broth cultures of these three toxigenic V. cholerae strains were admixed together after inactivation (with heat or formalin) at final concentration of 1011 CFU ml−1 (Dukoral monograph 2010; ATCC code: J07AE01).

Rabbit immunization

Ten female adult New Zealand white rabbits with average 2·5–3 kg weight were purchased from Pasteur Institute of Iran. The immunization procedures were carried out according to the animal care facility of Pasteur Institute of Iran. Rabbits were divided into five groups, naive group received sterile normal saline as negative control, and other groups were immunized with different antigens (rCTB, WC, WC-rCTB and WC-cCTB) as described in Table 2. Rabbits were immunized by the intraintestinal route as no additional chemical compound was used to protect antigens from the stomach enzymes and to guarantee that antigens can reach their targets on intestinal epithelial cells (Liang et al. 2003). Rabbits in each group were fasted for 24 h before surgery and anesthetized with proper dosage of ketamine and xylazine after which abdominal skin was shaved and sterilized with alcohol. Abdominal cavity was opened by vertical incision, and the ileocecal region was found. The end of this region was ligated to the inner wall of abdomen. In each group, the certain antigen was injected into proximal ileum, and the abdominal cavity was closed by stitching. The suture that tied the ileocecal region to the abdominal wall was removed 2 h after surgery, and rabbits were given water and food. Serum samples were collected from rabbits prior to immunization and in days 6, 9, 14, 21 and 28 after immunization (Liang et al. 2003).

Table 2. Concentration of antigens in each test groups
Group nameAntigen received
Naive 1 ml sterile normal saline
rCTB1 ml sterile normal saline containing 25 μg produced recombinant CTB
WC1 ml suspension of three inactivated Vibrio cholerae strains (1011 CFU ml−1)
WC-rCTB1 ml suspension of three inactivated V. cholerae strains plus 25 μg rCTB
WC-cCTB1 ml suspension of three inactivated V. cholerae strains plus 25 μg cCTB

Serum anti-CTB IgG antibody titre in immunized rabbits

To evaluate the maximum level of anti-CTB IgG titre in different days after immunization, the sera of rabbits were diluted 1 : 100 with PBS and GM1-ELISA was performed. Duplicate plates were used, one plate was coated with 1 μg per well cCTB, and the other plate was coated with 1 μg per well rCTB (rabbit anti-CT IgG was used as a positive control). To obtain the anti-CTB IgG titre in immunized rabbits, sera of different groups on day 28 were diluted three times serially in PBS from 1 : 100 to 1 : 218 700 and same GM1-ELISA was performed on diluted sera.

Rabbit ileal loop assay

The immunized rabbits were challenged with three different virulent V. cholerae strains (Table 1). Twenty-eight days after single-dose immunization, rabbits were starved for 24 h but given water. Surgery was performed after intramuscular anaesthesia. Abdomen of rabbits was opened, and their ileums were tied into 4-cm-long loops. One millilitre of different concentrations of challenge strains (3 × 105, 3 × 106, 3 × 107 and 3 × 10CFU ml−1) was injected in different loops (Liang et al. 2003). The abdomens were closed, and rabbits were given water and sacrificed after 18–20 h. Accumulated fluid in each loop was collected and measured. The fluid accumulation (FA) index was calculated from the ratio of loop fluid volume (ml) to loop length (cm) (Thungapathra et al. 1999).


Purification of recombinant CTB

In pAE-ctxB plasmid, the rCTB is produced with 6XHis-tag at the N-terminus and can be purified through Ni2+-charged column chromatography. Inclusion bodies were easily solubilized by urea, after which successfully refolded and purified by Ni2+-charged column chromatography. SDS-PAGE and Western blot analysis of eluted proteins showed a single 14-kDa band of purified rCTB (Fig. 1a,b,c).

Figure 1.

CTB protein purified with Ni2+-charged column chromatography. (a) Lane 1, protein size marker; lanes 2–11, different fractions of purified proteins through chromatographic column visualized with 15% SDS-PAGE. (b) SDS-PAGE analysis of purified rCTB in comparison with cCTB. Lane 1, rCTB and lane 2, cCTB. (c) Western blotting analysis of purified rCTB in comparison with cCTB. Lane 1, rCTB and lane 2, cCTB.

Concentration of eluted rCTB was evaluated with Bradford assay and was indicated to be approximately 700 μg ml−1.

Ganglioside GM1–enzyme-linked immunosorbent assay (GM1-ELISA)

GM1-ELISA functional assay was performed to confirm the ability of rCTB to bind to its cellular GM1 ganglioside receptor on epithelial cells. The results indicated that rCTB was able to bind to GM1 in a comparable manner with cCTB (data not shown).

Anti-CTB IgG titre in immunized rabbits

Indirect ELISA was performed to evaluate the ability of rCTB to produce anti-CTB IgG in immunized rabbits; anti-CTB IgG in different groups was assessed and subjected to statistical analysis (spss, ver. 17, IBM Corp., Armonk, NY, USA) with Games–Howell test. Antibody assays showed the slow increase in anti-CTB IgG antibodies after immunization, the peak level of which became detectable on day 28. There were no anti-CTB antibodies in the pre-immune sera. The sera of rabbits on day 28 postimmunization in microplate coated with rCTB showed that there is no significant disparity of IgG titre between groups of rabbits immunized by rCTB or WC-rCTB (P > 0·05) and it was approximately 1 : 8100 and 1 : 2700 for WC alone and WC-cCTB groups (P > 0·05). Rabbits immunized with whole cell in combination with rCTB or cCTB produced much higher level of IgG in contrast with group received only inactivated whole-cell V. cholerae (P < 0·05). Our results demonstrated some statistically significant differences between antibody levels when 96-well ELISA plate was coated with rCTB or cCTB (Fig. 2a,b). The IgG level in rCTB group and WC-rCTB was much higher than that of other groups when 96-well microplate was coated with rCTB (P < 0·05) (Fig. 2c), and when the plate was coated with cCTB, WC-cCTB group showed the highest level of anti-CTB IgG (P < 0·05) (Fig. 2d).

Figure 2.

Average anti-CTB IgG level in sera of different animal groups in 6 through 28 days: the 96-well microplate was coated with rCTB with cut-off point of about 0·155 (a) and cCTB with cut-off point of about 0·153 (b). Average titre of IgG on day 28 after immunization: the 96-well microplate was coated with rCTB, and anti-CTB IgG titre was approximately 1 : 8100 for rCTB and WC-rCTB groups and 1 : 2700 for WC and WC-cCTB groups (c); the 96-well microplate was coated with cCTB, and anti-CTB IgG titre was 1 : 8100 for WC-cCTB and WC-rCTB groups and 1 : 2700 for rCTB and WC groups (d); IgG titre in unimmunized rabbits was 1 : 900 in both plates (cut-off point of data was about 0·073). (image_n/jam12043-gra-0001.png) Normal saline; (image_n/jam12043-gra-0001.png) rCTB; (image_n/jam12043-gra-0001.png) WC-rCTB; (image_n/jam12043-gra-0001.png) WC and (image_n/jam12043-gra-0001.png) WC-cCTB.

Protection against challenges with toxigenic Vibrio cholerae strains

Fluid accumulation in different groups of rabbits was calculated, and the assigned FA values are depicted in (Fig. 3). Fluid accumulation index of ≥ 0·9 means that there is no effective immunity against challenged V. cholerae strains (Thungapathra et al.1999). The results showed that immunized rabbits in four test groups were protected against different doses of challenge strains, and this protection was higher in rCTB and rCTB-WC groups. In all groups, the higher challenge strains' concentration, the more fluid accumulated in loops was seen. In unimmunized rabbits, all loops showed significant amount of FA.

Figure 3.

Average fluid accumulation in immunized rabbit ileal loops after challenge with toxigenic Vibrio cholerae strains. (image_n/jam12043-gra-0001.png) rCTB; (image_n/jam12043-gra-0001.png) WC-rCTB; (image_n/jam12043-gra-0001.png) WC-cCTB; (image_n/jam12043-gra-0001.png) WC and (image_n/jam12043-gra-0001.png) normal saline.


Regional cholera outbreaks from 2005 through 2009 in Iran were due to Inaba strains that have been developed from their prevailing Ogawa (before 2005) by point mutation in O-antigen-encoding cluster (wbeT) (Sharifnia et al. 2012). Furthermore, sequence analysis revealed the presence of two ctxB genotypes among V. cholerae strains isolated from recent outbreaks (2004–2009) (Aliabad et al. 2012). This proposed the hypothesis of the circulation of inhabitant strains in the region, which further underlines the significance of the evaluation of immunological responses against a domestic rCTB-WC mixture. Our previously published data have indicated that different V. cholerae genotypes affected Iran in recent outbreaks (Bakhshi et al. 2008a). These genotypes were different from those reported in US and Europe, while identical genotypes were detected with those reported from Asian countries and Middle East (Bakhshi and Pourshafie 2009). This emphasizes on the need to develop a native vaccine composed of our local strains with their own surface structures and antigenic determinants.

This is now believed that CTB is effective oral immunogen that can stimulate mucosal immunity and create protection against cholera and enterotoxigenic E. coli infections (Sanchez and Holmgren 1986). According to the different studies by investigators, it has been proven that CTB is a potent immunogen on nasal and mucosal system (Sun et al. 1994; Rudin et al. 1999; Arêas et al. 2004; D'Ambrosio et al. 2008). Peru-15pCTB, attempt 1.3 (VA1.3) and IEM108 were examples of vaccines developed against cholera through the genetic manipulation of domestic or standard strains to produce higher amounts of CTB and reduced zero level of CTA (Thungapathra et al. 1999; Fontana et al. 2001; Liang et al. 2003; Roland et al. 2007).

In this study, rabbits were preferred to mice because rabbits have bigger body; thus, surgery would performed easier, and animals have more chance to be survived through experiment period. As described previously, rCTB was inoculated into rabbits intraintestinally; thus, we could guarantee that rCTB entered into the intestine of rabbit. The results showed that anti-CTB IgG was produced and gradually increased in all immunized groups. Serum anti-CTB IgG analysis demonstrates that inactivated toxigenic V. cholerae strains can produce higher level of IgG when inoculated into rabbit in combination with rCTB or cCTB (P < 0·05), this means that CTB can enhance the immune responses against V. cholerae probably via binding to its GM1 receptor at the surface of epithelial cells. In microplate that was coated with rCTB, the highest peak of IgG was related to group inoculated with rCTB and showed significant disparity with WC-cCTB group (P < 0·05); conversely, much higher level of anti-CTB IgG was seen in sera of rabbits inoculated with WC-cCTB in cCTB-coated microplate (P < 0·05). These results suggest that secreted IgG against rCTB has more binding affinity to rCTB than cCTB and vice versa.

The IgG titre against CTB, obtained from ELISA assay, was approximately 1 : 8100 in groups immunized with whole-cell V. cholerae plus rCTB or cCTB, which is threefold higher than IgG titre in rabbits immunized with inactivated whole-cell V. cholerae (1 : 2700), this means that rCTB increased the level of secreted IgG when used in combination with whole-cell V. cholerae; furthermore, rCTB, in separate, also produced proper level of anti-CTB IgG in serum. FA index was calculated in animal challenge test with live toxigenic V. cholerae O1 strains of both biotypes (classic and El Tor) and two serogroups (Ogawa and Inaba) of Iranian isolates. Recombinant CTB group was more protective (less FA index) against experimental infection with different concentrations of live toxigenic V. cholerae strains than other immunized groups, while FA index in other groups was still less than the naive group. Rabbits immunized with rCTB produced more anti-CTB IgG antibody than WC-rCTB group, which proposes the hypothesis that existence of other bacterial antigens in WC-rCTB construct applies part of immune system potential and causes a decrease in anti-CTB IgG level in this group. Furthermore, less anti-CTB IgG produced in WC-rCTB group results in less occupation of CTB fragments produced by challenge strains, and consequently, less protection and more fluid accumulate are obvious in ileal loops of rabbits in challenge experiment.

In conclusion, rCTB produced in this system showed biological and functional properties comparable with cCTB and could stimulate anti-CTB IgG responses in the serum of immunized rabbits. Besides, when rabbits were immunized with rCTB in combination with inactivated Iranian strains, they were protected against live toxigenic V. cholerae strains, made it a potential candidate for an indigenous vaccine.


The study was funded by Iranian National Science Foundation grant no. 90007471.