Protective immunity by oral immunization with heat-killed Shigella strains in a guinea pig colitis model

Authors


ABSTRACT

The protective efficacy of and immune response to heat-killed cells of monovalent and hexavalent mixtures of six serogroups/serotypes of Shigella strains (Shigella dysenteriae 1, Shigella flexneri 2a, S. flexneri 3a, S. flexneri 6, Shigella boydii 4, and Shigella sonnei) were examined in a guinea pig colitis model. A monovalent or hexavalent mixture containing 1 × 107 of each serogroup/serotype of heat-killed Shigella cells was administered orally on Days 0, 7, 14 and 21. On Day 28, the immunized animals were challenged rectally with 1 × 109 live virulent cells of each of the six Shigella serogroups/serotypes. In all immunized groups, significant levels of protection were observed after these challenges. The serum titers of IgG and IgA against the lipopolysaccharide of each of the six Shigella serogroups/serotypes increased exponential during the course of immunization. High IgA titers against the lipopolysaccharide of each of the six Shigella serogroups/serotypes were also observed in intestinal lavage fluid from all immunized animals. These data indicate that a hexavalent mixture of heat-killed cells of the six Shigella serogroups/serotypes studied would be a possible broad-spectrum candidate vaccine against shigellosis.

List of Abbreviations
cfu

colony forming unit

IgA

immunoglobulin A

IgG

immunoglobulin G

LPS

lipopolysaccharide

MOPS

3-[N-morpholino] propanesulfonic acid

OD

optical density

S.

Shigella

SD

standard deviation

TSA

tryptic soy agar

TSB

tryptic soy broth

Shigellosis remains a major public health problem worldwide, especially in pediatric subjects aged between 1 and 5 years in developing countries [1, 2]. The annual global incidence of shigellosis is about 164.7 million; 163.2 million of these patients are from developing countries. Moreover, approximately 1.1 million deaths are reported each year [3].

Shigellosis is caused by any of the following four serogroups of Shigella strains: S. dysenteriae, S. flexneri, S. boydii and S sonnei. The serogroups that cause shigellosis differ with time and geographical location. For example, S. sonnei is endemic in the developed world [3] whereas S. flexneri continues to be the most predominant in Asia and Africa [4-6]. During 1995–2000, a hospital-based study in Kolkata reported that S. flexneri was the predominant serogroup followed by S. sonnei, S. boydii and S. dysenteriae [7]. This trend was confirmed by recent active surveillance of acute diarrheal diseases in hospitalized patients in Kolkata [8, 9].

At present, antibiotic therapy only is recommended for the clinical management of shigellosis [10]. As a result, multidrug-resistant strains have emerged, creating a significant global health problem [10, 11]. Considering all recent events, the World Health Organization prioritized an area for development of a safe and effective vaccine for shigellosis [3, 12]. Although many candidate Shigella vaccines have been formulated [1, 13], there are still no licensed vaccines available.

We examined an animal model and report here the protective efficacy of heat-killed Shigella strains of various serogroups/serotypes, including S. dysenteriae 1, S. flexneri 2a, S. flexneri 3a, S. flexneri 6, S. boydii 4 and S. sonnei, as monovalent or multivalent immunogens.

MATERIALS AND METHODS

Bacterial strains

The strains of S. dysenteriae 1 (NT4907), S. flexneri 2a (B294), S. flexneri 3a (C519), S. flexneri 6 (C347), S. boydii 4 (BCH612), and S. sonnei phase I (IDH00968) used in this study were stock cultures from our laboratory. These strains were isolated from diarrheal patients admitted to the Infectious Disease Hospital in Kolkata. The invasiveness of these strains was confirmed by the Sereny test [14]. A Shiga toxin gene-deleted mutant of S. dysenteriae 1 (NT4907Δstx) was constructed as described previously [15].

Animals

Two-month-old English colored guinea pigs of both sexes, weighing 250–300 g, were supplied by the Animal Resource Department, National Institute of Cholera and Enteric Diseases, Kolkata, India. The Institutional Animal Ethical Committee approved the detailed protocols of the animal experiments (Approval Number: Appro/70).

Preparation of immunogen

Heat-killed cells of the Shigella strains were prepared as follows. Overnight cultures of S. dysenteriae 1 (NT4907Δstx), S. flexneri 2a (B294), S. flexneri 3a (C519), S. flexneri 6 (C347), S. boydii 4 (BCH612), and S. sonnei (IDH00968) were scraped off TSA (Difco, Sparks, MD, USA) after incubation at 37°C. The bacterial cells were suspended in PBS, pH 7.4, and centrifuged at 10,000 g for 10 min. The resulting pellets were washed twice and resuspended in PBS to a specific OD600 value equivalent to 1 × 107 cfu/mL. The suspensions were heated at 100°C for 1 hr, washed twice by centrifugation and resuspended in PBS. The resulting suspension was adjusted to the same OD600 as the suspension before heating (1 × 107 cfu/mL) and used as a monovalent immunogen. To prepare a hexavalent immunogen mixture, bacterial suspensions (6 × 107 cfu/mL in PBS) of the six Shigella serogroups/serotypes were prepared separately as described above and heated at 100°C for 1 hr. Washing, resuspension in PBS and adjustment of the concentration were carried out as described above. Equal volumes of the six immunogens were thoroughly mixed and stored at −80°C until use. An aliquot of this mixture was tested for viability by plating on TSA followed by overnight incubation at 37°C.

Preparation of challenge organism

Log-phase cultures in TSB (Difco) at 37°C of each of the six serogroups/serotypes of the Shigella strains were centrifuged, resuspended in PBS to a specific OD600 value equivlent to 1 × 109 cfu/mL and used immediately for challenge.

Oral immunization with heat-killed cells of the Shigella strains

Oral immunization of guinea pigs with 1 mL of the heat-killed cell suspensions of the Shigella strains was carried out on Days 0, 7, 14, and 21, as described previously [15]. Briefly, the animals were fasted for 36 hr but given water ad libitum. Before immunization, each guinea pig was anesthetized by an i.m. injection of a mixture of Ketamine (35 mg/kg body weight; Sterfil Laboratories, Mumbai, India) and Xylazine (5 mg/kg body weight; AstraZeneca Pharma, Kolkata, India). Five minutes later, Ranitidine (50 mg/kg body weight; Ranbaxy, Haryana, India) was administered i.v., followed by two administrations of 5 mL of 5% sodium bicarbonate solution directly into the stomach through a sterile saline-lubricated human infant feeding tube (Romsons, Bengaluru, India) at 15 min intervals. Immediately after the second administration of sodium bicarbonate solution, 1 mL of either monovalent or hexavalent heat-killed bacterial suspension was orally administered. The animals were then returned to their cages and given limited amounts of sterile water and food. In the control group, the animals received the same treatments as the immunized animals except for administration of 1 mL of sterile PBS instead of the immunogens.

Rectal challenge with virulent Shigella strains

Rectal challenge was carried out as described previously [16] with some modifications [15]. After four successive oral immunizations, the guinea pigs in the immunized and control groups were challenged on Day 28 with cells of a virulent Shigella strain. Before the challenge, each animal was sedated with a single i.m. injection of a mixture of Ketamine (35 mg/kg body weight) and Xylazine (5 mg/kg body weight). For rectal challenge, a single-use sterilized human infant feeding tube (approximately 40 cm in length and 4 mm in diameter; Romsons) was used, about 15–20 cm of the tube being inserted into the animals. A 1 mL aliquot containing 1 × 109 cfu of bacterial cells in PBS was given through the tube. After 15 min, the tube was carefully removed, thus avoiding backflow of the inoculum. After recovery from sedation, all animals were given access to sterile water and food.

The challenged animals were observed at 12 and 24 hr after the challenge for their level of general activity, tenesmus, consistency of stools passed into the drop pans of their cages, and presence of blood or mucus in the stools. Body weight and rectal temperatures were also measured.

Collection of stool samples and bacterial quantification

The animals' stool samples were collected from the drop pans and suspended in PBS at 1 g stool/mL. Bacterial quantification was performed on Hektoen enteric agar (Difco) and MacConkey agar plates (Difco). Representative colonies were subjected to confirmation with appropriate Shigella serotyping antisera (Denka Seiken, Tokyo, Japan).

Collection of serum samples

Blood samples were collected from the animals' footpad veins. The serum was separated and stored at −20°C until use.

Collection of intestinal lavage

Intestinal lavages were collected from the animals as described previously [15]. Briefly, 20–25 cm of small intestine was removed (mainly the jejunal portion) and 2 mL of PBS containing 0.1% BSA, 50 mM EDTA and 0.1 mg/mL soybean trypsin inhibitor was passed through and collected. Phenylmethylsulfonyl fluoride was added to the intestinal wash after collection, followed by vigorous vortexing and centrifugation at 10,000 g for 10 min to remove intestinal tissue cells and debris. The supernatants were collected, sodium azide (0.1%, w/v, final concentration) added and the mixture stored at −20°C until use.

Collection of intestinal tissue and assessment of bacterial colonization

Laparotomies were performed and portions of the animals' intestine (3–4 cm, distal colon) excised, minced, mixed with 3 mL of PBS, and homogenized with a pestle (HiMedia, Mumbai, India). The homogenized tissue was adjusted to 5 mL with PBS. Bacterial quantification and strain confirmation with appropriate antisera were performed as described above.

Preparation of lipopolysaccharide

Lipopolysaccharide was extracted from Shigella cells as described previously [15]. Briefly, each strain was cultured at 37°C overnight in TSB (100 mL) with constant shaking. The cells were then washed with distilled water and resuspended in 20 mL of 150 mM NaCl containing 20 mM MOPS, pH 6.9. An equal volume of buffer (20 mM MOPS, pH 6.9)-saturated phenol was added to the mixture and incubated at 65°C for 30 min with occasional shaking. The mixture was kept on ice for 10 min and then centrifuged at 15,000 g for 20 min. The top aqueous phase was collected, with four volumes of chilled ethanol added, and the mixture kept overnight at −20°C. The precipitated LPS was collected by centrifugation at 10,000g for 10 min. The carbohydrate content of the LPS was estimated by the phenol–sulfuric acid method.

Enzyme-linked immunosorbent assay

Analyses of anti-LPS serum IgG and IgA and of anti-LPS mucosal IgA were carried out by ELISA as described previously [15]. Briefly, each well of a disposable polystyrene microtiter plate (Nunc, Roskilde, Denmark) was coated with 1 µg of LPS of the immunizing strain in 100 µL of PBS. Control wells were coated with 100 µL of PBS. A horseradish peroxidase-conjugated goat anti-guinea pig IgG (Sigma Chemical, St. Louis, MO, USA) or horseradish peroxidase-conjugated sheep anti-guinea pig IgA (ICL, Portland, OR, USA) was used for detection of IgG and IgA, respectively. After completion of the reaction, the resulting color was measured by the absorbance at 492 nm using an ELISA reader (Bio-Rad, Hercules, CA, USA). The readings of the control wells were subtracted from those of the corresponding test wells to yield ODs of 0.100 or greater.

Statistical analysis

Statistically analyzed data are presented as the mean ± SD. P values were calculated by Student's t-test. Values of P ≤ 0.05 and P < 0.001 were considered to be statistically significant and highly significant, respectively.

RESULTS

Protective efficacy of heat-killed cells of Shigella strains

Table 1 shows the protective efficacy of oral immunization of guinea pigs with heat-killed cells of the Shigella strains under study against rectal challenge with live, virulent cells of these strains. The upper half of this table shows the protective efficacy conferred by oral immunization with heat-killed cells of a single serogroup/serotype of Shigella strain (monovalent immunogen) against rectal challenge with the homologous Shigella serogroup/serotype. In two series of experiments, each animal of six groups of 10 animals (60 in all) was orally administered four doses of 1 × 107 cfu of heat-killed cells of either S. dysenteriae 1 (NT4907Δstx), S. flexneri 2a (B294), S. flexneri 3a (C519), S. flexneri 6 (C347), S. boydii 4 (BCH612) or S. sonnei (IDH00968) in 1 mL of PBS on Days 0, 7, 14 and 21 (immunized group). In addition, 60 control animals (10 animals per group) each received 1 mL of PBS. On Day 28 (7 days after the last immunization), all animals were challenged with 1 × 109 cfu of the Shigella strains of the homologous serogroups/serotypes. No dysenteric symptoms were observed in any of the animals in the immunized groups, whereas all animals in the control groups developed symptoms of bacillary dysentery, such as tenesmus, mucoidal bloody diarrhea and bloody diarrhea. Tenesmus began to occur within 24 hr of challenge in all control animals. Mucoidal bloody diarrhea and mucoidal diarrhea without blood also occurred within 24 hr of challenge. Furthermore, increased rectal temperatures (1.6–2°C) and loss of body weight (10–12%) were observed in the control animals within 24 hr of challenge. No such symptoms developed in any of the animals in the immunized groups (Table 2).

Table 1. Protective efficacy of heat-killed cells of Shigella strains
ImmunogenExperimental groupStrain for rectal challengeNo. of animals examinedNo. of animals with symptoms
TenesmusMucoidal bloody diarrheaMucoidal diarrhea
  • Data from two experiments.
  • Observations were made 24 hr after rectal challenge.
  • Control groups were administered PBS.
S. dysenteriae 1ImmunizedS. dysenteriae 110000
 Control 101064
S. flexneri 2aImmunizedS. flexneri 2a10000
 Control 101046
S. flexneri 3aImmunizedS. flexneri 3a10000
 Control 101037
S. flexneri 6ImmunizingS. flexneri 610000
 Control 101046
S. boydii 4ImmunizedS. boydii 410000
 Control 101037
S. sonneiImmunizedS. sonnei10000
 Control 101019
HexavalentImmunizedS. dysenteriae 18000
 Control 8853
HexavalentImmunizedS. flexneri 2a8101
 Control 8844
HexavalentImmunizedS. flexneri 3a8000
 Control 8835
HexavalentImmunizedS. flexneri 68000
 Control 8826
HexavalentImmunizedS. boydii 48000
 Control 8817
HexavalentImmunizedS. sonnei8101
 Control 8817
Table 2. Rectal temperatures and body weights in animals after rectal challenge with live Shigella strains
ImmunogenExperimental groupStrain for rectal challengeNo. of animals examinedRectal temperature (°C; mean ± SD)Body weight (g; mean ± SD)
Before challengeAfter challengeBefore challengeAfter challenge
  • Data from two experiments.
  • Examinations were made 24 hr after challenge. Immunized versus control groups, P < 0.05.
  • §Control groups were administered PBS.
  • One immunized animal with tenesmus and mucoidal diarrhea was excluded.
S. dysenteriae 1ImmunizedS. dysenteriae 11038.19 ± 0.638.24 ± 0.4431 ± 20434 ± 20
 Control§ 1038.3 ± 0.540.2 ± 0.3426 ± 26375 ± 22
S. flexneri 2aImmunizedS. flexneri 2a1038.1 ± 0.1738.5 ± 0.55437 ± 16442 ± 16
 Control 1038.4 ± 0.440.21 ± 0.4425 ± 18379 ± 17
S. flexneri 3aImmunizedS. flexneri 3a1038.4 ± 0.3138.7 ± 0.38435 ± 5.8442 ± 6
 Control 1038.2 ± 0.2240 ± 0.18410 ± 13368 ± 11
S. flexneri 6ImmunizedS. flexneri 61038.8 ± 0.3738.9 ± 0.36428 ± 14435 ± 13
 Control 1038.5 ± 0.840.5 ± 0.41418 ± 21371 ± 19
S. boydii 4ImmunizedS. boydii 41038.71 ± 0.438.73 ± 0.4412 ± 11418 ± 11
 Control 1038.3 ± 0.340.1 ± 0.5410 ± 10370 ± 19
S. sonneiImmunizedS. sonnei1038.5 ± 0.438.58 ± 0.5424 ± 20429 ± 21
 Control 1038.4 ± 0.3840 ± 0.19423 ± 15378 ± 13
HexavalentImmunizedS. dysenteriae 1838.2 ± 0.538.4 ± 0.55437 ± 32441 ± 32
 Control 838.4 ± 0.640.7 ± 0.5410 ± 46359 ± 40
HexavalentImmunizedS. flexneri 2a738.3 ± 0.3838.4 ± 0.41421 ± 12425 ± 11
 Control 838.6 ± 0.4940.2 ± 0.47419 ± 17372 ± 15
HexavalentImmunizedS. flexneri 3a838.6 ± 0.3638.8 ± 0.3416 ± 14420 ± 15
 Control 838.7 ± 0.2940.7 ± 0.4415 ± 9377 ± 6
HexavalentImmunizedS. flexneri 6838.4 ± 0.4238.7 ± 0.26405 ± 19410 ± 19
 Control 838.3 ± 0.4339.8 ± 0.5411 ± 12374 ± 11
HexavalentImmunizedS. boydii 4838.5 ± 0.438.6 ± 0.37421 ± 11427 ± 11
 Control 838.4 ± 0.4740.3 ± 0.44418 ± 13371 ± 12
HexavalentImmunizedS. sonnei738.4 ± 0.4538.8 ± 0.28424 ± 15429 ± 15
 Control 838.5 ± 0.3640.3 ± 0.42432 ± 27385 ± 24

The lower half of Table 1 shows the protective efficacy of oral immunization with heat-killed cells of a mixture of the six serogroups/serotypes (hexavalent immunogen) of the Shigella strains against rectal challenge with virulent cells of one of the six serogroups/serotypes of the Shigella strains. In two series of experiments, each animal of six groups of 8 animals (48 animals in all) was orally administered four doses of 1 mL of the hexavalent immunogen mixture containing 1 × 107 Shigella cells of each of the six serogroups/serotypes on Days 0, 7, 14 and 21 (immunized group). The control animals (48 animals; 8 per group) each received 1 mL of PBS orally. On Day 28 (7 days after the last administration), the animals in each of the immunized and control groups were challenged with 1 × 109 cfu of virulent cells of Shigella strain of either S. dysenteriae 1 (NT4907), S. flexneri 2a (B294), S. flexneri 3a (C519), S. flexneri 6 (C347), S. boydii 4 (BCH612) or S. sonnei (IDH00968). Most animals in the immunized groups developed no dysentery symptoms within 24 hr of challenge. However, one animal challenged with S. flexneri 2a (B294) and one challenged with S. sonnei (IDH00968) developed symptoms of dysentery, namely tenesmus and mucoidal diarrhea (Table 3). On the other hand, all control animals developed typical symptoms of dysentery with tenesmus and bloody or mucoidal diarrhea. Significant body weight loss (9–12%) and increased rectal temperatures (1.5–2.3°C) were observed in the animals in the control groups (Table 2).

Table 3. Clinical responses of symptomatic guinea pigs in hexavalent immunized groups
ImmunogenExperimental groupStrain for rectal challengeNo. of animal with symptomsNo. of Shigella (cfu/g of stool)No. of Shigella (cfu/g of distal colon)Anti-LPS IgA titerRectal temperature (°C)Body weight (g)
Before challengeAfter challengeBefore challengeAfter challenge
  • Examinations were made 24 hr after challenge.
  • Data are shown as reciprocal log2 titers.
HexavalentImmunizedS. flexneri 2a19 × 1032 × 10663840.1431396
  S. sonnei11.4 × 1041 × 106538.439.8401364

Recovery of challenge strains from the stools and distal colons of the challenged animals

At 24 hr after challenge, the animals' stools were examined for bacterial cells; the results are shown in Table 4. In the upper half of Table 4, the results for monovalent immunization are presented. Large numbers of bacterial cells of the challenge strains ([7.5 ± 1.9] × 104 to [4.2 ± 3.2 × 106]) were recovered from the animals in the control groups. On the other hand, significantly fewer bacterial cells were recovered from the animals in the immunized groups (18.2 ± 11.1 to 50.2 ± 33.5) than were recovered from the corresponding control groups.

Table 4. Recovery of the challenge Shigella strains from the stools of the challenged animals after immunization with heat-killed cells
ImmunogenExperimental groupStrain for rectal challengeNo. of animal examinedNo. of Shigella (cfu/g of stool; mean ± SD)
  • Data from two experiments.
  • Examinations were made 24 hr after challenge. Immunized versus control groups, P < 0.05.
  • §Control groups were administered PBS.
  • One immunized animal with tenesmus and mucoidal diarrhea was excluded.
S. dysenteriae 1ImmunizedS. dysenteriae 1524.4 ± 14.5
 Control§ 5(4.2 ± 3.2) × 106
S. flexneri 2aImmunizedS. flexneri 2a545.2 ± 23
 Control 5(3.0 ± 3.5) × 106
S. flexneri 3aImmunizedS. flexneri 3a518.2 ± 11.1
 Control 5(3.6 ± 3.7) × 105
S. flexneri 6ImmunizedS. flexneri 6529.0 ± 12.3
 Control 5(1.7 ± 2.9) × 106
S. boydii 4ImmunizedS. boydii 4520.2 ± 8.9
 Control 5(3.9 ± 2.8) × 105
S. sonneiImmunizedS. sonnei550.2 ± 33.5
 Control 5(7.5 ± 1.9) × 104
HexavalentImmunizedS. dysenteriae 1418.7 ± 10.8
 Control 4(5.0 ± 2.4) × 106
HexavalentImmunizedS. flexneri 2a350.6 ± 25.1
 Control 4(3.5 ± 2.7) × 106
HexavalentImmunizedS. flexneri 3a429.8 ± 17.7
 Control 4(5.1 ± 2.5) × 105
HexavalentImmunizedS. flexneri 6414.3 ± 7.8
 Control 4(9.0 ± 3.3) × 105
HexavalentImmunizedS. boydii 4417.0 ± 7.8
 Control 4(4.6 ± 5) × 105
HexavalentImmunizedS. sonnei355.3 ± 21.0
 Control 4(9.0 ± 0.89) × 104

In the lower half of Table 4, the results for recovery of challenge strains from the stools of animals immunized with the hexavalent immunogen mixture are presented. Large numbers of bacterial cells of the challenge strains ([9.0 ± 0.89] × 104 to [5.0 ± 2.4]) × 106) were recovered from the animals in the control groups. On the other hand, fewer of the challenge bacteria (14.3 ± 7.8 to 55.3 ± 21.0) were recovered from the majority of animals in the immunized groups. However, 9 × 103 and 1.4 × 104 cfu/g of stool of the challenge strains were recovered from the stools of the two animals that had developed symptoms (the one challenged with S. flexneri 2a and the one challenged with S. sonnei, respectively) (Table 3).

At 24 hr after challenge, bacterial cells of the challenge strains recovered from the distal colons of the animals were examined. In the experiments on monovalent immunization, large numbers of bacterial cells of the challenge strains ([1.7 ± 2.4) × 108 to [2.6 ± 3.4] × 109) were recovered from the animals in the control groups. On the other hand, significantly fewer bacterial cells (16.1 ± 8.71 to 28.7 ± 22.8) were recovered from the animals in the immunized groups than were recovered from the corresponding control groups. In the experiments on hexavalent immunization, significantly large numbers of bacterial cells of the challenge strains ([4.9 ± 6.8) × 108 to [6.1 ± 13.7] × 109) were recovered from the animals in the control groups. On the other hand, significantly fewer of the challenge bacteria (11.1 ± 2.6 to 15.2 ± 8.27) were recovered from the majority of animals in the immunized groups. However, 2 × 106 and 1 × 106 cfu/g of the challenge strains were recovered from the distal colons of the two animals that developed symptoms (the one challenged with S. flexneri 2a and the one challenged with S. sonnei, respectively).

The significantly lower recovery of the challenge strains from the stools and distal colons of the immunized animals clearly indicate that the challenged strains could not colonize these animals' intestines.

Serum immunoglobulin G and immunoglobulin A titers against lipopolysaccharides of the immunized strains

Serum IgG and IgA titers against LPS of the Shigella strains of the animals in both the immunized and control groups were quantified and the results are shown in Fig. 1a,b, respectively. In the immunized groups, both anti-LPS serum IgG and IgA titers against the immunized strains increased during the period of immunization (up to Day 21). The increases in titers were sustained during the 7 days after the last immunization on Day 21, especially the anti-LPS IgA titers. In the control groups, both IgG and IgA titers were below the detection limits during the entire experimental period.

Figure 1.

Anti-LPS serum IgG and IgA titers after monovalent immunization. Blood samples were collected from the animals' footpads on Days 7, 14, 21 and 28 and serum samples prepared. (a) Serum IgG titers and (b) IgA titers are shown. Each bar represents the mean ± SD of five determinations. Immunized versus control groups P < 0.001. ○, S. dysenteriae 1; ●, S. flexneri 2a; □, S. flexneri 3a; ▪, S. flexneri 6; Δ, S. boydii 4; ▴, S. sonnei; ♦, control.

Serum IgG and IgA titers against LPS of the Shigella strains were also measured after immunization with hexavalent immunogen mixture. The serum IgG titers against the LPS of all six Shigella strains in the immunized groups increased during the period of immunization (up to Day 21; Fig. 2a). As shown in Fig. 2b, the serum anti-LPS IgA titers against all six Shigella strains in the immunized groups also continued to increase during the period of immunization and after immunization (up to Day 28; Fig. 2b). In the control groups, both the IgG and IgA titers were below the detection limits during the entire experimental period (Fig. 2a,b).

Figure 2.

Anti-LPS serum IgG and IgA titers after hexavalent immunization. Serum (a) IgG titers and (b) IgA titers are shown. Each bar represents the mean ± SD of four determinations except for the groups receiving S. flexneri 2a and S. sonnei, from each of which one animal with tenesmus and mucoidal diarrhea was excluded. Immunized versus control groups P < 0.001. ○, S. dysenteriae 1; ●, S. flexneri 2a; □, S. flexneri 3a; ▪, S. flexneri 6; Δ, S. boydii 4; ▴, S. sonnei; ♦, control.

Immunoglobulin A titers against lipopolysaccharides of the immunized strains in intestinal lavage

At 24 hr after challenge, the anti-LPS IgA titers of the intestinal lavage fluid of the animals in the immunized and control groups were measured and the results are shown in Table 5. As shown in the upper half of this table, the titers of all groups immunized with a single serogroup/serotype of Shigella strains were significantly higher than those of the control groups (P < 0.001 in all six groups). As shown in the lower half of Table 5, after hexavalent immunization the IgA titers against the LPS of each serogroup/serotype of the Shigella strains were significantly higher than those in the control groups (P < 0.001 in all six groups).

Table 5. Anti-IgA titers in the intestinal lavages of the animals immunized with heat-killed cells of the Shigella strains
ImmunogenStrains examined for IgA titerNo. of animal examinedAnti-LPS IgA titer
Immunized groupControl group§
  • Data from two experiments.
  • Examinations were made 24 hr after challenge. Immunized versus control groups, P < 0.001.
  • §Control groups were administered PBS.
  • Data are expressed as means ± SD of the reciprocal log2 titers.
  • ††One immunized animal with tenesmus and mucoidal diarrhea was excluded.
S. dysenteriae 1S. dysenteriae 157.84 ± 5.833.14 ± 2.13
S. flexneri 2aS. flexneri 2a56.14 ± 5.132.14 ± 1.13
S. flexneri 3aS. flexneri 3a56.85 ± 4.843.20 ± 1.84
S. flexneri 6S. flexneri 6510.26 ± 8.844.27 ± 2.84
S. boydii 4S. boydii 456.20 ± 5.722.68 ± 1.13
S. sonneiS. sonnei56.14 ± 5.132.38 ± 1.42
HexavalentS. dysenteriae 148.58 ± 7.23.32 ± 2
HexavalentS. flexneri 2a3††7 ± 02.32 ± 1
HexavalentS. flexneri 3a47.32 ± 63 ± 0
HexavalentS. flexneri 6410.8 ± 94 ± 0
HexavalentS. boydii 446.58 ± 5.22.58 ± 1.2
HexavalentS. sonnei3††6.41 ± 5.22.32 ± 1

DISCUSSION

The leading candidate vaccines for shigellosis thus far are polysaccharide conjugates, synthetic conjugates, ribosomal subunits, invasion complex-based, outer membrane vesicles and live attenuated vaccines [2, 13, 17, 18]. Oral administration of inactivated whole cells is another strategy for developing an effective vaccine against shigellosis [19]. Potential advantages of inactivated whole-cell vaccines are that their preparation is relatively easy and inexpensive and their administration does not require needles, thereby making them practical for use in developing countries. There is only limited evidence for inactivated whole-cell vaccines against shigellosis. It has previously been shown that acetone-killed Shigella is an effective mucosal immunogen [20]. An orally administered formalin-inactivated whole-cell vaccine for S. sonnei was effective in a double-blind, placebo-controlled phase I trial [21]. These reports are in contrast to an earlier study in which monkeys were protected by oral vaccination with attenuated organisms, but not by an acetone-inactivated vaccine [22]. We have previously reported that heat-killed S. flexneri 2a protects against challenge with the homologous strain in a rabbit model [23]. In that model, five doses of 1 × 1011 bacteria were orally administered to rabbits at 7-day intervals and the immunized animals challenged 7 days after the last dose of vaccine. Homologous protective efficacy was 100% and highly significant. A further study showed that protection could be achieved with a minimum of four doses, whereas fewer doses did not give protection against challenge in rabbit and guinea pig ocular models [24].

In the present study, we used a guinea pig colitis model that has already been proven useful for assessing the protective efficacy of Shigella vaccine candidates [15, 16]. Based on reported data concerning prevalence of serogroups/serotypes that cause shigellosis globally as well as locally in Kolkata, we selected S. dysenteriae 1, S. flexneri 2a, S. flexneri 3a, S. flexneri 6, S. boydii 4 and S. sonnei [1, 2, 4, 5, 8, 9]. Heat-killed cells of single serogroup/serotype of Shigella strains as both monovalent and hexavalent immunogens induced significant protective immune responses; however, further studies are needed to elucidate the mechanisms underlying such protection. Several studies have shown that anti-LPS antibodies are elicited in response to Shigella infection, both locally as secretory IgA and systematically as serum IgG [25, 26]. In our study, serum IgG and IgA titers against the LPS from each of the six serogroups/serotypes increased during four successive immunizations with heat-killed cell suspensions. Serum IgG can potentially neutralize pathogens that enter the mucosa and thus prevent systemic spread [27]. The strong serum IgG responses in the immunized guinea pigs may have provided robust resistance to invasion by the challenge Shigellae by enhancing phagocytic clearance of the organisms; this would be a possible mechanism by which protection against bloody diarrhea is conferred. We also found high anti-LPS IgA titers against each of the six immunized strains in intestinal lavage fluid from the guinea pigs. Gut-derived IgA responses are also thought to have a significant role in mucosal defense [1, 27]. However, after immunization with the hexavalent immunogen mixture, one of eight guinea-pigs (12.5%) was not protected against challenge by a virulent strain of either S. flexneri 2a or S. sonnei. With the monovalent vaccination, 1 × 107 cfu of the Shigella strains were enough to induce protective immunity; however, the same amount of the bacteria may not be enough in hexavalent vaccination. This controversial finding needs to be further investigated.

Despite the promising results of this study, several other issues remain to be addressed including: (i) whether the current strategy of using a multivalent heat-killed vaccine can generate long-term protective efficacy; (ii) whether it will possible to induce high levels of immunogenicity with lower doses of immunogens; (iii) whether it will be possible to obtain sufficient cross-protective immunogenicity with an appropriate combination of different serogroups/serotypes; and (iv) whether there will be a limit to the number of serogroup/serotype that can be included in a multivalent immunogen. Further studies to elucidate these issues are in progress in our laboratory.

ACKNOWLEDGMENTS

This study was supported by the Japan Initiative for Global Research Network on Infectious Diseases (J-GRID), Ministry of Education, Culture, Sports, Science and Technology, Japan. SB was a fellow sponsored by J-GRID.

DISCLOSURE

No authors have any conflict of interests to disclose.

Ancillary