AI-2 quorum sensing controlled delivery of cytolysin-A by tryptophan auxotrophic low-endotoxic Salmonella and its anticancer effects in CT26 mice with colon cancer

Abstract

Introduction: The limitations of conventional cancer therapies necessitate target-oriented, highly invasive, and safe treatment approaches. Hence, the intrinsic anti-tumor activity of Salmonella can offer better options to combat cancers.

Objectives: This study aims to utilize attenuated Salmonella and deliver cytolytic protein cytolysin A (ClyA) under quorum sensing (QS) signaling for precise localized expression in tumors but not in healthy organs.

Methods: The therapeutic delivery strain was imposed with tryptophan auxotroph for selective colonization in tumors by trpA and trpE deletion, and lipid-A and O-antigen were altered by pagL and rfaL deletions using lambda red recombination method. The strain was transformed with the designed QS-controlled ClyA expression vector which was validated by western blot. The in vivo passaged therapeutic strain was used for treatment four times at a weekly interval, with a dose of 5 × 10⁶ CFU/mouse for cancer therapy.

Results: The attenuated strain induced minimal endotoxicity-related cytokines TNF-α, IL-1β, and IFN-γ and exhibited adequate colonization despite earlier exposure in mice. The QS-controlled ClyA expression was confirmed by western blot from bacterial cultures grown at different cell densities. The results demonstrated that the in vivo passaged strain preferentially colonized the tumor after vacating the spleen, liver, and lung, leaving no outward histological scars. The anti-cancer effect of the designed construct was evaluated in the murine CT26 colon cancer model. The expression of ClyA increased tumoricidal activity by 67 % compared to vector control.

Conclusion: Hence, the anti-tumor effect of the engineered Salmonella strain was improved by ClyA expression via QS activation after achieving the threshold bacterial cell density. Further, immunohistochemical staining of the tumor and other organs corroborated the QS-controlled tumor-specific expression of ClyA. Overall, the results imply that the developed anti-cancer Salmonella has low endotoxicity and QS-controlled expression of ClyA as beneficial safety elements and supports recurrent Salmonella inoculation by O-antigen deficiency.

Keywords: Cytolysin A; Quorum sensing; Reduced endotoxicity; Salmonella; Tryptophan auxotroph; Tumoricidal activity.

Graphical abstract

Fig. 1

Schematic experimental layout of the development of the therapeautic strain and its evaluation for anti-tumor effects.

Fig. 2

Cloning and in vitro evaluation of cytolysin A expression under the quorum sensing (QS) promoter. A. pJHL90 was constructed by replacing the P_trc promoter in pJHL65 with the QS promoter. The clyA gene was inserted in the MCS region of pJHL90 using restriction enzymes, BamHI and SalI. B. Growth curve. Overnight grown cultures of WT Salmonella Typhimurium (ST) and attenuated ST JOL2955 were sub-cultured in LB broth. Bacterial growth was measured by optical density (OD) at 600 nm every hour for 10 h. C. Expression of cytolysin A protein under QS promoter. The bacterial lysate prepared at 0.5, 1.5, 1.8, and 2.0 OD₆₀₀ values was evaluated for ClyA expression by western blot. M = protein marker and +ve = positive control, cytolysin A protein extracted from E. coli BL21 (DE3). D. Hemolytic activity on mice erythrocytes. The suspension of 2 % erythrocytes was incubated for 4 h with protein extracts and centrifuged to detect hemolysis. Hemolysis by JOL2955 was evaluated at OD₆₀₀ values of 0.5 and 1.5. The tubes labeled C and VC denote control and vector control, respectively.

Fig. 3

Attenuated Salmonella influences bacterial colonization and impairs endotoxicity. A. Effect of priming on the colonization of Salmonella. Salmonella infection was ascertained in the presence of pre-existing antibodies induced by priming intraperitoneally with O-antigen-positive (JOL2952) and -negative (JOL2953) strains (1 × 10⁶ CFU/mice). Primed mice were injected with O-antigen-positive and O-antigen-deficient (JOL2953) Salmonella strains. Mice were sacrificed at 3 days post inoculation to enumerate bacterial CFU in the spleen and liver at primed and non-primed states. Data were analyzed by unpaired t-test. B. The expression of inflammatory cytokines, TNF-α, IL-1β, and IFN-γ. Cytokine expression was measured by RT-PCR in the spleen collected at 3 days post-inoculation. Data were analyzed by Tukey’s multiple comparison tests using one-way ANOVA. The error bars denote the SEM. ns = not significant, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. D. Histopathological examination of the spleen and liver. Hematoxylin and eosin staining was performed to analyze the effect of Salmonella infection in the organs. The investigation showed prominent tissue dispersion of red pulp with diffused lining between red and white pulp indicated by the black arrow and infiltration of inflammatory cells denoted by the red arrow in the spleen infected with wildtype (WT) and JOL2953. Infiltration of macrophages in the liver is indicated by the green circle. Noticeably altered tissue architecture was not demonstrated between JOL2954 and the PBS-inoculated group. Images were obtained at 200× magnification. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

Evaluation of adhesion, invasion, and intracellular survival and the cytolysin A effect. A. JOL2955 was passaged in vivo in CT26 tumor-bearing mice thrice to obtain JOL2955P3 and the strain was compared to its original, JOL2955P0, for adhesion, invasion, and intracellular survival evaluation in CT26 and HT29 cells. Data were analyzed by the unpaired t-test. *p < 0.05 and **p < 0.01 where the error bars denote the SEM. B. Wound healing assay. A confluent monolayer of CT26 cells was scratched in the middle to create a gap between cells. Cells were washed and treated with JOL2955P3, vector control (VC), or media alone as a control (C) for 2.5 h. Extracellular bacteria were eliminated by applying gentamycin in the culture media. Cells monitored for 48 h revealed the gap closure in C and VC groups, as denoted by the black circle. Cell proliferation was inhibited in the scratched regions in the treated group. C. Apoptosis induced by the strains. CT26 cells were treated overnight with JOL2955P3 and VC. Cells were stained with propidium iodide (PI) and annexin V-FITC and evaluated by a fluorescence-activated cell sorting (FACS) reader to differentiate live and apoptotic cells. D. Histogram of apoptosis induced by attenuated strains. The cells were categorized into live (PI^- annexin V^-), early apoptosis (PI^- annexin V⁺), and late apoptosis (PI⁺ annexin V⁺) based on their stage of apoptosis corresponding to staining. Data were analyzed by Tukey’s multiple comparison tests using one-way ANOVA, *p < 0.05, **p < 0.01, and ****p < 0.0001, where the error bars denote the SEM.

Fig. 5

Safety assessment of the attenuated Salmonella Typhimurium JOL2955P3 strain. A. Fitness of JOL2955 in the tumor after three successive in vivo passages in CT26 tumor-bearing mice. The syngeneic CT26 BALB/c mice were inoculated with JOL2955, where the strain was recovered from the tumor and subsequently passaged to obtain JOL2955P3. The colonization of the passaged strain was compared with that of its parental strain in the tumors of CT26 mice at 3, 7, and 14 days post-infection (dpi) using the unpaired t-test. B. Survival curve for BALB/c mice challenged with 5 × 10⁷ CFU/mL by an intraperitoneal route (IP). C. Colonization of Salmonella Typhimurium JOL2955P3 in the tumor, spleen, liver, and lung. Tumor-bearing mice were inoculated with 5 × 10⁶ CFU/mL bacteria via the IP route. Five mice were sacrificed at each time point. The organs were harvested, homogenized, and plated in BGA plates to enumerate the bacterial colonies that invaded the organs. Data were analyzed by Tukey’s multiple comparison tests, one-way ANOVA, *p < 0.05 and ****p < 0.0001, where the error bars denote the SEM. D. Histopathological changes in the spleen, liver, and lung. The organs were harvested 2 weeks after the final dose and fixed in paraffin. The tissue sections were deparaffinized and stained with hematoxylin and eosin for histopathological analysis. The effect on the organs was investigated and compared with organ samples of mice receiving PBS as a control. No notable changes were investigated in the organs of challenged mice. Images were obtained at 100 × magnification.

Fig. 6

In vivo anti-tumor effects of JOL2955P3. A. Schematic schedule for the cancer treatment using Salmonella Typhimurium JOL2955P3. The tumor was induced in six-week-old BALB/c mice with subcutaneous (SC) injection of 5 × 10⁶ cells/mL CT26 cells. Animals were challenged intraperitoneally with the strains, and the tumor volume was measured by a Vernier caliper for up to 35 days post-treatment. B. Gross morphological appearance of the tumors in mice sacrificed at the end of the experiment. The red circle denotes the position and size of the tumor. C. Gross size and appearance of tumors and spleens harvested from respective groups. D. The volume (V) of a tumor was measured by applying the tumor volume formula, V = ½ (length × width²). The weights of the E. tumors and F. spleens. Mice were sacrificed at the end of the experiment, and the tumors and spleens were harvested and weighed. Data were analyzed by Tukey’s multiple comparison tests, one-way ANOVA, ns = not significant, *p < 0.05, **p < 0.01, and ***p < 0.001, where the error bars denote the SEM. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 7

Effects of treatment in tumors. A. Expression of cytokines and cell cycle regulatory genes. Relative in vivo expression of A. cytokines and B. apoptosis-related genes. RNA was extracted from the tumors and the expression of cytokines and apoptosis-related genes was quantified by RT-PCR. B. Expression of tumor-associated macrophage (TAM) markers in tumors. TAMs were isolated from tumors by density gradient centrifugation using Ficoll-Paque plus density gradient media. Isolated TAMs were stained with CD68-PE, iNOS-FITC, and CCL2-APC and quantified by FACS. A higher number of marker-positive cells was noted in tumors from non-treated mice, while positive cells were fewer in tumors from mice treated with JOL2955P3. Data were analyzed by Tukey’s multiple comparison tests, one-way ANOVA, *p < 0.05, **p < 0.01, and ***p < 0.001, where the error bars denote the SEM. C. Hematoxylin and eosin (H&E) staining of tumors. Paraffin-fixed tumor tissues were sectioned, deparaffinized, and stained with H&E. The necrotic lesions in treated samples are denoted by red arrows. D. Immunohistochemical (IHC) observation of tumors, spleens, and livers to confirm the expression of cytolysin A (ClyA). Tissue sections were treated with anti-ClyA rabbit polyclonal antibodies and developed color with DAB. The brown spots marked inside the circle indicate the expression of ClyA in the tumor region of treated samples. The absence of brown spots in the spleen and liver in treated mice denotes the controlled expression of ClyA. The samples from the vector control and placebo groups also lack brown spots. Images were acquired at 200× for H&E and at 400× magnification for IHC. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)