Bacterial strain and culture conditions

KWIK-STIK of S. flexneri 2b serotype ATCC 12022 (Microbiologics, USA) was used in this study. For dissolving the bacterial strain, each stik was punched to release the hydrating fluid and mixed with lyophilized bacteria. ATCC 12022 was cultivated by the streak plate technique on deoxycholate citrate (DCA) (Oxoid, UK) agar under aerobic conditions at 37 °C for 24–48 h [47]. (30%) Luria–Bertani (LB) (Lab M, Neogen company, UK) glycerol stock was prepared for storing isolated colonies of S. flexneri 2b serotype ATCC 12022 at -80 °C for further use. S. flexneri 2b serotype ATCC 12022 was used for producing BGCs, as a control for protein and DNA measurements, and as a bacterial inoculum in the challenge experiment.

Minimum inhibitory concentration (MIC) determination of “TX100” against ATCC 12022

Following the recommendations of the American Society for Microbiology, (TX100) (ADVENT, CHEMBIO PVT.LTD) was tested for its (MIC) against S. flexneri 2b serotype ATCC 12022 using the broth dilution method [48]. In short, S. flexneri 2b serotype ATCC 12022 inoculum was cultivated overnight in 5 mL LB at 37 °C, as 150 µL was transferred to 15 mL of fresh LB broth and incubated at 37 °C in a shaker incubator at 200 rpm for 3 h, the bacterial inoculum was adjusted at (1 × 106 CFU/mL, OD600:0.2). Different dilutions of 10 mL of v/v TX100 were prepared (1, 2, 3, 4, 5, 6, 7, 8, 9, and 10%), and 100 µL of bacterial inoculum was added and incubated at 37 °C for 24 h. At last, a turbidity inspection was conducted in means of optical density (OD) determination at 600 nm for all incubated tubes, and the MIC value was recorded. The lowest TX100 concentration that showed no visible turbidity (complete bacterial killing) was recorded as the MIC value, and sub-MIC concentrations directly below the MIC value were used for BGC production. The sub-MIC concentration was tested at different incubation periods (1, 3, 6, 12 & 24 h) to obtain the ideal non-damaging incubation time for producing BGCs [18, 49].

Bacterial ghost cell (BGCs) production

In this study, two sets of BGCs were prepared through incubation with two different types of surfactants, namely (TW80) and (TX100), each of which was used to prepare an individual preparation.

(TW80) method for production of S. flexneri BGC (Sf-BGC/TW80)

Shigella flexneri bacterial ghost cells (Sf-BGCs) were produced using the (TW80) method as described by Rabea et al. with some modifications [18]. A 100 µL of bacterial inoculum ATCC 12022 (1 × 1010 CFU/mL, OD600:0.6) was added to 10 mL of v/v (7%) TW80/LB (Alpha Chemika), and that mixture was incubated for 24 h at 37 ℃. After incubation, 180 µL of lactic acid was added to the treated cells in order to get the pH down to 2.8, and 1 h of incubation was performed. BGCs were centrifuged for 10 min at 4000 rpm, and the supernatant was saved for future use. The treated cells were washed and centrifuged twice at 4000 rpm with (0.45%) NaCl. Finally, BGC pellets were suspended in (0.45%) NaCl and kept at −20 °C for further use.

(TX100) method for production of S. flexneri BGC (Sf-BGC/TX100)

This BGCs set was prepared using 10 mL of v/v (5%) TX100/LB using the same bacterial inoculum (1 × 1010 CFU/mL, OD600:0.6). That concentration of (TX100) was corresponding to the sub-MIC value of surfactant against the bacterial strain ATCC 12022 (5%). The exact TW80 protocol was followed except for reducing the incubation time with the surfactant to 12 h only.

For each BGC's production method, biological and technical triplicates were performed.

Quality check for the produced BGCsIn vitro quality check

The produced batches of BGCs went through a number of tests to ensure their quality in vitro. The presence of transmembrane tunnels in the cell wall of treated bacteria, negative growth upon re-cultivation, intactness, preservation of cellular integrity, and a substantial release of protein content and DNA from the treated cells in comparison to untreated native cells were assessed. All of the listed tests served as assurances of the successful production.

Light microscopic examination

For evaluating the cellular integrity and intactness of the treated cells, one loop from the preserved BGC pellets from each preparation was Gram-stained and visualized under the light microscope (× 1000).

BGCs re-cultivation

This step was performed to ensure that all living cells in the S. flexneri BGC pellets from each preparation had died during the manufacturing process. Samples from these pellets were sub-cultured for 24–48 h at 37 °C on DCA agar and LB broth. Colonies or turbidity (OD600) were observed, and any evidence of growth was noted.

Quantification of the released protein and DNA cellular contents

Using Nanodrop (Jenway-7415 NANO) at 280 nm and 260 nm, the released protein content and DNA were measured in the supernatant from the centrifuged BGCs, respectively [18, 50]. Quartz cuvette was used for both measurements, where an extinction E260 = 1 that corresponds to 50 μg dsDNAmL − 1 was used for DNA measurement. While for protein concentration measurements, Bovine Serum Albumin (BSA) standard curve was used [15]. Supernatants of centrifuged, native, untreated S. flexneri 2b serotype ATCC 12022 (1 × 1010 CFU/mL, OD600:0.6) cultured in LB for 12 h and 24 h at 37 ℃ served as controls for TX100 and TW80 preparations, respectively [50]. Three replicates of each BGC's preparation were used to measure the released protein and DNA contents. The average, standard deviation, and percentage of increase were computed.

Scanning electron microscopy (SEM) examination of “Sf-BGC”

Scanning electron microscopy (SEM Quanta FEG) was used to examine the morphological features (transmembrane tunnels) of the produced BGCs [16]. Cells were fixed for 2 h at room temperature in (2.5%) glutaraldehyde (SIGMA-Aldrich) in 0.1 M phosphate buffer (pH 7.0), washed three times with Phosphate-buffered saline (PBS), and then post-fixed for 1 h at room temperature in (1%) osmium tetroxide (SIGMA-Aldrich). For dehydrating the fixed cells, several grades of ethanol concentration (10, 30, 50, 70, and 100%) and liquid CO2 were used. Finally, after coating the cells with gold-palladium and mounting them on SEM stubs, they were examined using an SEM Quanta FEG with a 20 kV accelerating voltage and a x16000 amplification power. The identification of transmembrane tunnels on the surface of bacterial cells guaranteed the successful development of BGCs.

In vivo quality check “stool cultures”

Using an in vivo model, the viability of Sf-BGCs was also examined by culturing stool samples from experimental animals that had been immunized with the ghost cells on a selective medium (DCA). This step was conducted to ensure the safety of the used BGC preparation in the absence of living S. flexneri. Stool samples were taken before and after each dose of the bacterial ghost vaccines (BGVs), as well as before and after the S. flexneri 2b serotype ATCC 12022 (IP) challenge. Such samples were obtained using the procedures outlined in [51], with some modifications. Briefly, a fresh stool sample was collected by placing each mouse in a clean cage, where filter papers were placed underneath the mouse. A stool sample was collected using a sterile inoculating loop, suspended in 1 mL of sterile PBS, and vortexed for 1 min. Each sample was centrifuged for 1 min at maximum speed, and the supernatant was used for the viable count experiment. Briefly, 20 µL of each sample supernatant was serially diluted in 180 µL of sterile, cold PBS. Finally, 10 µL of each dilution was plated on DCA media and incubated for 24 h at 37 °C. Viable, colorless colonies of S. flexneri were counted, and CFU/mL was calculated for each plate using the following equation: CFU/mL=((No.of colonies x diltution factor))/(Volume plated)‎.

$$}/} = \frac}.\,}\,}\, \times \,}\,})}}}\,}}}$$

Immunogenicity evaluation of BGVs

Animals were housed at a constant temperature of 25 °C in a 12 h light/ 12 h dark cycle with free access to regular pellet food and water. The mice were provided with a week of accommodations before the evaluation began. Before starting the immunogenicity evaluation experiment, a pilot study of (IP) infection with ATCC 12022 was carried out to ensure the successful infection in BALB/C mice. Six adult male BALB/C mice (6 weeks old at the day of the experiment) were divided equally into two groups, three in each, as follows: negative control (non-immunized received 20 µL of sterile PBS (IP), positive control (non-immunized infected once with S. flexneri 2b serotype ATCC 12022 (IP)). The progress of the infection was assessed through post-infection stool cultures in addition to monitoring the survival of the subjects for seven days. After a full assessment of the intraperitoneal infection in mouse subjects, the immunogenicity evaluation of BGVs began. Twenty-eight adult male BALB/C mice (6 weeks old on the day of the experiment) were divided equally into four groups, seven in each, as follows: negative control (non-immunized received sterile PBS), positive control (non-immunized infected with S. flexneri 2b serotype ATCC 12022), Sf- BGV/TW80 (immunized with BGC prepared with TW80 then challenged with S. flexneri 2b serotype ATCC 12022), and Sf-BGV/TX100 (immunized with BGC prepared with TX100 then challenged with S. flexneri 2b serotype ATCC 12022). Mice were first immunized with the BGVs to test their safety, and then they were challenged with the (IP) infection of S. flexneri 2b serotype ATCC 12022.

Immunization with BGVs

The (IP) route of immunization was used three times at intervals of 2 weeks [52, 53]. The TW80 and TX100 BGV groups received 20 µL (1 × 108cells/mL) from each ghost preparation, while the negative control group received 20 µL of sterile PBS. To verify the lack of living S. flexneri cells and determine the safety of each preparation, stool samples were obtained from each mouse and plated on DCA media for counting viable CFU at 1, 3, 6, 12, and 24 h post-immunization with BGVs. Also, pathological scores of stool samples were assessed after each of the immunization doses of BGVs (1, 3, 6, 12, and 24 h post-immunization) as described in [53], color (brown, yellow, light green, blue-green), consistency (normal, loose, soft, hard), and an accumulative number of diarrheal episodes during the day (0–3).

Challenge experiment

The pure bacterial inoculum of S. flexneri 2b serotype ATCC 12022 was adjusted to (1 × 108 CFU/dose) at OD600:0.4 for the challenge experiment, which began 2 weeks after the last immunization dosage (week 7). The actual bacterial inoculum was plated by serial dilution on DCA media prior to dose administration to confirm the infection and incubated for 24 h at 37 ℃. All groups received (IP) challenge dosage of S. flexneri 2b serotype ATCC 12022 20 µL once, except for the negative control, which received 20 µL of sterile PBS. Pathological scores of shigellosis were observed in all tested groups, as described in [53]. Also, stool samples were collected from all groups (1, 3, 6, 12, and 24 h post-challenge), plated on DCA media, and viable CFUs were counted.

Blood collection and serum harvesting

After the completion of the immunization and challenge experiment (24 h post-challenge) for all groups under anesthesia, the retro-orbital sinus puncture blood collection method was carried out and termination procedures as described in [54, 55]. Sera were extracted from blood samples after coagulation (20–30 min) and centrifugation at 6000 rpm for 10 min. All samples were kept in sterile 1 mL microcentrifuge tubes and stored at −20 °C for further use.

Bacteriological analysis

According to Vinod and his colleagues, the protective efficacy and safety of Sf-BGV preparations were evaluated against the virulently challenged group (positive control) [16]. At 24 h post-challenge, all of the experimental mice were sacrificed. Since the intestinal infection (shigellosis) was the focus of the immunization and challenge experiment, 0.25 cm of the colon was aseptically collected and intestinal content was evacuated and flashed with sterile, cold PBS. The collected tissues were manually homogenized in 1 mL of cold, sterile PBS. After being homogenized, all of the tissues were serially diluted ten times and then placed on DCA agar for 24 h at 37 °C. In terms of CFU/mL, the bacterial count for each culture plate was calculated.

Determination of serum bactericidal activity assay (SBA) of “Sf-BGV”

Serum sample from animal groups were collected at the end of immunization and challenge experiment (24 h post-challenge). The procedure for determining serum bactericidal activity (SBA) is outlined in [16, 56, 57]. In short, 100 µL of S. flexneri 2b serotype ATCC 12022 adjusted at (1 × 106 CFU/mL, OD600:0.2) and added to 25 µL of sera from immunized and non-immunized challenged groups. After incubating for 1 h at 37 °C, samples were cultured on DCA agar for 24 h at 37 °C. Serum was replaced with 25 µL of sterile PBS in the PBS/control tube. Using the following equation, the SBA% was determined:

$$}\% \, = \,1\, - \,\frac}\,}\,}\,}\,}\,}\,}} \right)}}}\,}\,}\,}\,}\,}\,}} \right)}}\, \times \,100\%$$

Detection of specific humoral immune response (IgG) by Enzyme-Linked Immunosorbent assay (ELISA)

At the end of the immunization and challenge experiment (24 h post-challenge), all of the mice's sera were examined to determine the humoral immune response (IgG). A volume of 50 µL (1 × 109 CFU/mL≈ OD600:0.5–0.6) of S. flexneri ATCC 12022 was used for coating 96-well microtiter ELISA plates for serum samples obtained from Sf-BGV/TW80, Sf-BGV/TX100, and the positive control group. The bacterial harvest was prepared by overnight culturing the bacterial strain in LB. The bacterial load was collected by centrifugation at 12,000 rpm for 5 min and the bacterial precipitates were sonicated in an ice bath for 5 min to break down the cells. The sonicated cells were centrifuged as previously mentioned, and the supernatants were collected. The protein content of cell lysates was determined using the Lawry method and was used to coat the ELISA plates using 1 µg/mL [58]. On the other hand, wells used for negative control samples were coated with irrelevant polypeptide obtained from Streptococcus pyogenes ATCC 19615 (Microbiologics, USA) that was cultivated in Tryptic Soy Broth (TSB) (Oxoid, UK) in order to test antibody specificity as a "mock vaccine”, where the same method of the coating was used [19, 22]. All plates were blocked using (1%) blocking buffer of bovine serum albumin (BSA) (SIGMA-Aldrich) for 1 h. Sera samples were twofold serially diluted and applied at 0.1 mL/well for 1 h at 37 °C. Serum dilutions were decanted, and plates were washed out using 200 µL of wash buffer. Goat anti-mouse IgG HRP per-oxidase conjugate (0.5 mg/mL) (Elabscience Biotechnology Inc.) was used as 0.1 mL of a 1/1000 final dilution. Plates were incubated for 1 h and washed as previously. All plates were treated with 100 µL of 3,3′,5,5′-Tetramethylbenzidine (TMB) substrate (Elabscience Biotechnology Inc.), and the produced color was stopped with 2N H2SO4. All experimental groups' mean optical densities were measured using microplate reader at 450 nm. To obtain the exact titer endpoint, sera were diluted twofold, and the highest dilution that gave an absorbance greater than three times that of the wells that received all treatments except the sera (the negative control) was used in the calculations.

Histopathology analysis of colon tissues

From each of the tested groups, samples of the colon's tissue were taken. After flushing and being fixed for 72 h in (10%) neutral buffered formalin (Carolina Biological Supply Co.) samples were trimmed, processed in serial grades of ethanol, cleared in xylene, and infiltrated and embedded in Paraplast wax tissue embedding media (SIGMA-Aldrich). A rotatory microtome was used to cut 4-micron (µ)-thick sections of tissue that were mounted on glass slides from various samples. A Full HD microscopic imaging system (Leica Microsystems GmbH, Germany) was used to examine tissue sections after they had been stained with Hematoxylin and Eosin as a typical general morphological examination stain. The scoring system for lesions was used as described in [59]; the scoring system was: "Nil", as there were no visible lesions; + A mild lesion was found in less than (15%) of the tissue that was examined; ++A moderate lesion was found in (16–35%) of the tissue sections examined; +++ A severe lesion was found in more than (35%) of the tissue that was examined. Damage to the lining epithelium, infiltrates of inflammatory cells, and the condition of the blood vessels (congested or dilated) were the goals of the analysis. For sample fixation and staining, all standard procedures were followed [60].

Statistical analysis

Statistical differences were analyzed using GraphPad Prism (8.0.2) (La Jolla, CA, USA) using one-way analysis of variance (ANOVA) and two-way analysis (ANOVA-mixed models); p-values with < 0.05 and 0.01 indicated significance and high significance, respectively.