Tel.: +886-4-2205-3366 ext: 2173, Fax: +886-4-2205-3764
A polymer coating applied to Salmonella prevents the binding of Salmonella-specific antibodies
Article first published online: 14 JUL 2012
Copyright © 2012 UICC
International Journal of Cancer
Volume 132, Issue 3, pages 717–725, 1 February 2013
How to Cite
Lee, C.-H., Lin, Y.-H., Hsieh, J.-L., Chen, M.-C. and Kuo, W.-L. (2013), A polymer coating applied to Salmonella prevents the binding of Salmonella-specific antibodies. Int. J. Cancer, 132: 717–725. doi: 10.1002/ijc.27700
- Issue published online: 26 NOV 2012
- Article first published online: 14 JUL 2012
- Accepted manuscript online: 26 JUN 2012 11:33PM EST
- Manuscript Accepted: 1 JUN 2012
- Manuscript Received: 20 SEP 2011
- National Science Council. Grant Numbers: NSC 100-2320-B-039-024, NSC 100-2815-C-039-031-B
- China Medical University. Grant Numbers: CMU99-N2-08, CMU100-N2-01
- poly(allylamine hydrochloride);
- neutralizing antibody;
The use of Salmonella as a potential antitumor agent has been investigated, but innate immunity against this bacterium reduces the efficacy of its tumor-targeting and antitumor activities. The purpose of this study was to investigate the modulation of the tumor-targeting efficiency of Salmonella enterica serovar choleraesuis by modifying the immune response to these bacteria by coating them with poly(allylamine hydrochloride) (PAH), designated PAH-S.C. To evaluate this modulation, we used naïve mice and mice immunized with Salmonella to study the role of the preexisting immune response to the antitumor activity of PAH-S.C. When anti-Salmonella antibodies were present, the invasion activity, cytotoxicity, and gene transfer of Salmonella was significantly decreased, both in vitro and in vivo. Treatment with PAH-S.C. resulted in delayed tumor growth and enhanced survival in immunized mice. Furthermore, immunohistochemical studies of the tumors revealed the infiltration of neutrophils and macrophages in immunized mice treated with PAH-S.C. These results indicate that Salmonella encapsulation effectively circumvented the Salmonella-specific immune response.
One of the primary limitations of cancer therapy is the lack of selectivity of therapeutic agents against tumor cells. The use of preferentially replicating bacteria as an oncolytic agent is an innovative approach for cancer treatment.1–3 This approach is based on the observation that several obligate or facultative anaerobic bacteria are capable of selectively multiplying in tumors and thereby inhibiting their growth.4 Salmonella are Gram-negative, facultative anaerobes that are a common cause of intestinal infections. Salmonella grow under aerobic and anaerobic conditions; thus, they are able to colonize small metastatic and larger tumors. Attenuated Salmonella have been shown to inhibit tumor growth in a broad range of human and mouse tumors.5
Salmonella-based vectors have been considered as potential antitumor agents for tumor vaccines, gene delivery and tumor-targeting vectors.6–8 However, a major disadvantage to this approach is the development of immunity to Salmonella.9 Salmonella-specific antibodies significantly reduce the tumor-targeting activity of Salmonella following a second administration of Salmonella. Humoral responses, particularly those involved in the production of neutralizing antibodies, reduce Salmonella's antitumor activity. Moreover, pre-existing Salmonella neutralizing antibodies are common in humans due to the ubiquitous nature of Salmonella serotypes10; therefore, these antibodies may interfere with Salmonella's antitumor activity even at the first inoculation of Salmonella. Masking Salmonella immunogenicity with a linking polymer is one approach that can be used to inhibit antibody-mediated Salmonella neutralization.
Purified proteins, plasmids, and viruses can be encapsulated in a biodegradable polymer and delivered by various routes.11 The interactions of polymers with biological systems has been a topic of interest in widely divergent fields. Living cells can be encapsulated by the alternating adsorption of oppositely charged polyelectrolytes, and the metabolic activity of the coated cells is well preserved after encapsulation.12, 13 The bacterial cell wall is negatively charged due to the presence of either teichoic acid in Gram-positive bacteria or lipopolysaccharide (LPS) in Gram-negative bacteria. Positively charged polyelectrolytes can bond to negatively charged bacterial surfaces to form thin films that circumvent preexisting immunity against bacterial vectors. In addition to rendering tumor-targeting Salmonella less susceptible to inactivation by neutralizing antibodies, this study demonstrates that the masking of Salmonella with a polymer reduces the antigenicity of Salmonella following in vivo delivery.
Material and Methods
Bacteria, cell lines, reagents and mice
The Salmonella enterica serovar choleraesuis (S. choleraesuis; S.C.) (ATCC 15480) vaccine strain was obtained from the Bioresources Collection and Research Center (Hsinchu, Taiwan). This rough variant of S.C., which is designated vaccine 51, was obtained by spreading an 18-hr broth culture of the virulent strain 188 of the S. choleraesuis serovar Dublin over the surface of a dried nutrient agar plate, adding a drop of a suspension of Salmonella anti-o phage No. 1, and selecting for a phage-resistant colony after incubation at 37°C for 24 hr.8, 9 S.C. were transformed with pTCYLuc by electroporation to generate S.C./Luc as previously described.4 Murine 4T1 breast tumor cells were cultured in Dulbecco's modified Eagle's medium supplemented with 50 μg/ml gentamicin, 2 mM L-glutamine and 10% heat-inactivated fetal bovine serum at 37°C in 5% CO2.14 Poly(allylamine hydrochloride) (PAH; MW: 15,000), 4′,6-diamidino-2-phenylindole (DAPI), fluorescein amine isomer I (FA) and paraformaldehyde were purchased from Sigma-Aldrich (St. Louis, MO). Six- to eight-week-old female BABL/c mice were obtained from the National Laboratory Animal Center of Taiwan. The animals were maintained in a specialized pathogen-free animal care facility in isothermal conditions with regular photoperiods. The experimental protocol adhered to the rules of the Animal Protection Act of Taiwan and was approved by the Laboratory Animal Care and Use Committee of the China Medical University (permit number: 99-20-N).
Preparation of PAH-modified Salmonella
S.C. [106 colony-forming units (cfu)] were washed with deionized water to remove nutrients and metabolites from the bacteria. The washed S.C. were dissolved in 1 ml of PAH solution (1.25–20 mg/ml), incubated on a shaker for 15 min and then centrifuged (10 min at 1,000g). The excess PAH solution was discarded, and the S.C. were dispersed and washed three times in water.
Characterization of the PAH-modified Salmonella
The size distributions and zeta potential values of the S.C. and the PAH-modified S.C. (PAH-S.C.) were measured in deionized water using a dynamic light scattering system (Zetasizer ZS90, Malvern instruments, Malvern, UK). The synthesis of FA was based on a reaction between the FA amine groups and the PAH carboxylic acid groups. PAH and FA were dissolved in acetonitrile and incubated at room temperature for 2 hr. To remove the unconjugated FA, the solution was dialyzed in the dark in distilled water, which was replaced daily until no fluorescence was detected in the supernatant. The resultant FA-PAH was lyophilized in a freeze dryer.15 The PAH-S.C and FA-PAH-S.C. cells were prepared as previously described. The PAH-S.C. and S.C. incubated 4T1 cells, respectively, for 2 hr. The cells were then washed with Phosphate buffered saline (PBS), fixed in 3.7% paraformaldehyde and observed by confocal laser scanning microscopy. The images were superimposed using the LCS Lite software. The FA-PAH-S.C. were collected at various time points before the measurements. The fluorescence emission from the solutions was measured using a fluorescence spectrometer (LS50B, PerkinElmer, Emeryville, CA).
Groups of mice were intraperitoneally (i.p.) immunized with heat-killed S.C. or PAH-S.C. at a dose of 104 cfu. The immunized and naïve mice were bled or sacrificed at several time points to determine antibody production.9 Other groups of naïve and immunized mice were subcutaneously (s.c.) inoculated with 106 tumor cells. When the tumors had grown to diameters between 50 and 100 mm3, the mice were intravenously (i.v.) injected with 5 mg/ml PAH or 2 × 106 cfu of S.C. or PAH-S.C. These groups of mice were sacrificed at various time points postinfection, and the numbers of Salmonella in the tumors, livers and spleens were determined on Luria broth (LB) agar plates; these data were expressed as cfu per gram of tissue. Tissue homogenates were also assessed for luciferase activity using a luciferase reporter gene assay system (Tropix). In a parallel experiment, the fluorescence emission from solutions was measured using a fluorescence plate reader (Tecan Safire II, Salzburg, Austria). In a separate experiment, palpable tumors were measured every 3 days in two perpendicular axes using a tissue caliper, and the tumor volumes were calculated as follows: (length of tumor) × (width of tumor)2 × 0.45. The survival rates of the mice in the treated and control groups were monitored daily.
Infection of tumor cells with Salmonella
S.C. (106 cfu) or PAH-S.C. (106 cfu) were incubated with various sera collected from mice and cultured with 4T1 cells (105 cells/well) for 8 hr. The medium was then removed, the cells were washed and fresh medium supplemented with 50 μg/ml gentamicin was added. The supernatants were removed after an additional 90 min culture in gentamicin, and adherent cells were lysed to release the intracellular bacteria. The lysates were serially diluted in PBS and spread on LB agar plates, and the cfu were counted after an overnight incubation at 37°C. In a parallel experiment, the adherent cells were measured for cell survival. Cell survival was assessed using the colorimetric WST-1 assay (Dojindo Labs, Tokyo, Japan) according to the manufacturer's instructions. Tumor cells (105/well) were cultured in six-well plates overnight. Subsequently, S.C./Luc or PAH-S.C./Luc admixed with various sera collected from mice were added to cells that were then cultured in 1 ml of antibiotic-free medium and incubated for 8 hr. All the cells were washed, replenished with complete medium containing gentamicin (50 μg/ml) and cultured for 16 hr. The cells were lysed to prepare extracts for the determination of luciferase activity using a luciferase assay kit (Tropix).
Assessment of cytokines and immunohistochemical staining
The levels of cytokines, tumor necrosis factor α (TNF-α) and interleukin-1 β (IL-1β) in the sera of the mice after S.C. or PAH-S.C. administration were determined by enzyme-linked immunosorbent assay (R□D, Minneapolis, MN). To analyze cell infiltration in the tumors, groups of mice that had been inoculated s.c. with 106 4T1 cells at day 0 were injected i.v. with 2 × 106 cfu of S.C. or PAH-S.C. at day 10, and the control mice received PBS. The tumors were excised and snap-frozen on day 20. Cryostat sections (5 μm) were prepared, fixed and incubated with rat anti-mouse Ly 6G (Gr-1) (RB6-8C5, BD Biosciences, San Diego, CA) or rat anti-mouse Mac-3 (M3/84, BD Biosciences) antibodies. After sequential incubation with the appropriate peroxidase-labeled secondary antibody and aminoethyl carbazole as the substrate chromogen, the slides were counterstained with hematoxylin. The numbers of infiltrating cells were quantified by averaging the number in each cell type in the three areas of highest cell density at 200× magnification in each section.15
The unpaired, two-tailed Student's t test was used to determine differences between groups for the comparison of tumor volume, Salmonella luciferase activity, cell number, body weight and cytokine production. A survival analysis was performed using the Kaplan-Meier survival curve and log-rank test. A p value less than 0.05 was considered to be statistically significant.
Characterization of PAH-modified Salmonella
Because the surface charge of the bacterial wall is negative, the positively charged polyelectrolyte PAH is effectively and spontaneously absorbed onto the bacterial cell wall.13 The particle sizes and surface charges of PAH-S.C. cells prepared with different PAH concentrations were measured; the results are listed in Table 1. Dynamic light scattering analysis demonstrated that the average size of the PAH-S.C. particles increased with increasing PAH concentration. The surface charges of the S.C. and PAH-S.C. particles were measured by zeta potential. The net surface charges of the PAH-S.C. and S.C. particles ranged from −4.07 to 11.75 mV. The gradual increase in particle size and neutralization of the Salmonella surface charge suggest that the Salmonella were well coated with PAH.
We also measured the replication rates of PAH-S.C. and S.C. cells to determine whether the PAH coating on the surface of Salmonella affected the physiology of the bacterium. When Salmonella were coated with 20 mg/ml PAH, the Salmonella growth curve was slightly decreased compared to the control (Fig. 1a). The growth rates of the other groups were not significantly different from that of the S.C. group. At 20 mg/ml, PAH may affect the growth of bacteria by inhibiting cell wall formation.
To test whether the PAH-S.C. cells were able to infect tumor cells and to determine their infection efficiency in the absence and presence of a neutralizing antibody, the gentamicin protection assay was used to measure the degree of infection of tumor cells by PAH-S.C. The bacterial invasion assay demonstrated that the invasion efficiency of PAH-S.C. decreased with increasing PAH concentration (Fig. 1b). The same phenomenon was observed in the presence of a naïve antibody. Interestingly, the PAH-S.C. prepared with 5 mg/ml PAH retained invasion activity in the presence of a neutralizing antibody. The PAH-S.C. prepared with 20 mg/ml PAH lost invasion activity and cytotoxicity, both in the absence and presence of a neutralizing antibody (Fig. 1c).
Previously, we demonstrated that Salmonella are effective gene transfer vectors in vitro and in vivo.4, 16 To test whether PAH-S.C. were able to transfer genes into tumor cells, luciferase expression was measured after infection with PAH-S.C. carrying a luciferase gene (PAH-S.C./Luc) (Fig. 1d). Tumor cells infected with PAH-S.C. prepared with 20 mg/ml PAH displayed no significant luciferase signals, either in the absence or presence of a neutralizing antibody. Notably, the expression of luciferase in PAH-S.C. in the 5 mg/ml group was observed both in the absence and presence of a neutralizing antibody. Our results demonstrated that the anti-Salmonella antibody did not inhibit PAH-S.C. invasion. Therefore, coating Salmonella with 20 mg/ml PAH partially decreased its infection ability, and coating Salmonella with 1.25 mg/ml PAH did not protect Salmonella from the neutralizing antibody. The PAH-S.C. cells prepared with 5 mg/ml PAH were, therefore, used in the subsequent analyses.
To further demonstrate that PAH did indeed adhere to the Salmonella surface, PAH was fluorescently labeled with FA to form FA-PAH. As illustrated in Figure 2, the Salmonella cells were well covered with FA-PHA (Fig. 2a). We hypothesized that the Salmonella PAH coating would be diluted with each subsequent cell division. Furthermore, to determine the kinetics of PAH release in vitro, we collected the bacteria and removed the supernatants at several time points. As expected, the level of PAH coating continuously decreased during bacterial division (Fig. 2b). Additionally, we measured the fluorescence of the tumor homogenate after FA-PAH-S.C. administration. The FA-PAH levels in the Salmonella recovered from the tumors were determined at various time points (Fig. 2c). The fluorescence in the tumors 48 hr after injecting the mice with FA-PAH-S.C. was ∼20-fold lower than that in the tumors at 12 hr. This is in contrast to the kinetics of bacterial accumulation in tumors. Our data support the hypothesis that the PAH-S.C. retained the ability to infect tumor cells in vitro and in vivo.
Tissue distributions of PAH-S.C. in naïve and immunized mice
To investigate the impact of immune responses on the tissue distribution of PAH-S.C., naïve and immunized mice bearing tumors were i.v. injected with PAH-S.C., and the amounts of Salmonella in the tumors, livers and spleens were determined at days 1 and 10. As shown in Figure 3a, systemically administered S.C. or PAH-S.C. preferentially accumulated within the tumors on day 1, at tumor-to-normal tissue accumulation ratios of 1,000–10,000:1. The amounts of S.C. and PAH-S.C. in the tumors remained at high levels over the 10 days (Fig. 3b). We previously obtained similar results demonstrating that the bacterial load in tumors was significantly reduced in mice immunized against Salmonella.9 These findings suggested that the immune response inhibited the accumulation of Salmonella in the tumor sites. Notably, the PAH-S.C. restored the tumor-targeting ability of Salmonella in immunized mice.
As shown in Figure 3b, 10 days after PAH-S.C. inoculation, the amounts of Salmonella in the tumors were approximately four orders of magnitude higher than those found in tumors derived from immunized mice treated with S.C. In addition, luciferase expression was readily observed in the tumor sites after PAH-S.C./Luc inoculation (Fig. 3c). The PAH-S.C. displayed tumor-targeting efficiency and gene transfer ability in the presence of a host with anti-Salmonella immunity. Our data demonstrated that PAH-S.C. circumvented the anti-Salmonella antibodies in vitro and in vivo.
Effects of PAH-modified Salmonella on cytokine induction in naïve mice
Because significant proinflammatory cytokine induction occurs with Salmonella infection, the administration of agents such as TNF-α, TNF-α neutralizing antibodies and IL-1 receptor antagonists may reduce the biological activity of these cytokines and their side effects.17 To examine the host inflammation response induced by PAH-S.C., immunized and naïve mice were treated with S.C. or PAH-S.C, and body weight was measured. As indicated in Figure 4a, naïve mice treated with S.C. had a 9% lower average body weight compared to naïve mice treated with PBS. In contrast, the body weights of mice treated with PAH-S.C. were not significantly decreased. After S.C. and PAH-S.C. treatment, the levels of inflammatory cytokines including IL-1β and TNF-α were measured in the sera. Regardless of whether the mice were naïve or immunized, the induction of inflammation cytokines (i.e., IL-1β and TNF-α) in mice treated with S.C. was increased by 1.5- to 4.5-fold compared to the induction by PAH-S.C. treatment (Figs. 4b and 4c). Taken together, these results suggest that PAH-S.C. had a greater tumor-targeting efficiency with potentially fewer side effects in the host than S.C.
Inhibition of tumor growth by PAH-modified Salmonella
Previously, we demonstrated that the presence of anti-Salmonella antibodies results in a noticeably lower total number of bacteria in tumor sites and decreases the antitumor effect of Salmonella.9 Therefore, the avoidance of neutralizing antibodies emerges as a key issue in the development of Salmonella therapeutic approaches. The antitumor effects of PAH-S.C. and S.C. were evaluated in terms of tumor growth and survival in naïve and immunized mice bearing tumors. S.C. and PAH-S.C. treatment both reduced tumor growth compared to PBS treatment in naïve mice. Interestingly, PAH-S.C. treatment significantly reduced tumor size compared to S.C. treatment in immunized mice (Fig. 5a). The mean tumor volume for immunized mice in the PAH-S.C.-treated group decreased by 48.72% compared to those in the S.C.-treated group. Figure 5b demonstrates that the survival of the immunized mice injected with PAH-S.C. was significantly prolonged compared to that of the mice injected with S.C. To further confirm that PAH protects against neutralizing antibodies, we used PAH-S.C. to immunize mice. The anti-PAH-S.C. antibodies did not inhibit the antitumor activity of S.C (Fig. 5c). Furthermore, PAH did not produce an antitumor response. These results are consistent with our previous report that demonstrated that the specificity of anti-Salmonella antibodies influences the tumor-targeting potential of Salmonella.9 Overall, PAH-S.C. significantly suppressed tumor growth and prolonged the survival of immunized mice.
Increase of infiltrating cells by PAH-S.C.
The antitumor properties of Salmonella may partially rely on the activation of host immunity.18–21 Neutrophil and macrophage infiltration in tumors from tumor-bearing mice inoculated with S.C., PAH-S.C. and PBS were analyzed at 10 days postinfection by immunohistochemistry. The results of the immunohistochemical staining are presented in Figure 6a. A few macrophages and neutrophils were detected in the tumors derived from immunized mice treated with S.C. Nevertheless, a notable increase in macrophage and neutrophil infiltration in the tumors was observed in naïve mice treated with PAH-S.C. and, in particular, in immunized mice treated with PAH-S.C. The numbers of infiltrating neutrophils and macrophages in the tumors treated with S.C. and PAH-S.C. were significantly increased compared to those in tumors treated with PBS (Figs. 5b and 5c). Notably, the number of infiltrating cells in tumors derived from immunized mice treated with PAH-S.C. was significantly increased compared to those treated with S.C. Taken together, these results reveal that PAH-S.C. cells enhanced immune cell infiltration in tumors in immunized mice. PAH-S.C. not only resisted the binding of neutralizing antibodies in peripheral blood but also replicated in tumor sites, which has been suggested to stimulate nonspecific antitumor immunity.
Systemic administration of Salmonella in tumor-bearing mice leads to its preferential accumulation in tumor sites and thus retards tumor growth. Salmonella can effectively eradicate primary and metastatic tumors including bone, prostate, breast, pancreas and sarcoma.22–25 Many studies suggest the clinical potential of bacterial treatment for critical metastatic tumor targets.26–29 Salmonella with an attenuated lipid A have been evaluated in a Phase I clinical trial, in which patients received Salmonella (VNP20009), which was rapidly cleared from the blood, and the majority of tumors had no detectable Salmonella colonization. Moreover, patients that had pre-existing anti-Salmonella antibodies had no detectable Salmonella colonization in the tumor sites after systemic administration.17 In agreement with a clinical study, our previous results indicated that higher anti-Salmonella antibody titers in the host resulted in lower amounts of Salmonella in tumor sites. Anti-Salmonella antibodies result in a noticeably lower total number of bacteria in tumor sites and decrease the antitumor effect of Salmonella. Therefore, the avoidance of neutralizing antibodies has emerged as a key issue in the development of Salmonella therapeutic approaches. In this study, we investigated the use of PAH to modify or shield Salmonella from the pre-existing immune response in a host.
For this purpose, we used commercially available charged polyelectrolytes, namely, PAH, to produce shells that cover living Salmonella. We presumed that coating Salmonella with such a film would not seriously affect its viability or invasion ability. A high concentration of PAH (20 mg/ml) may alter the bacterial outer membrane permeability by binding to the negatively charged LPS layer, resulting in a destabilized outer membrane that inhibits bacterial replication and invasion. Conversely, a low PAH concentration (1.25 mg/ml) was insufficient to block Salmonella antigenicity and protect against antibody neutralization.
Because toxic adverse effects caused by bacterial therapy are the main hindrance to further clinical application,16 we explored the use of PAH-S.C., which did not induce a large response of inflammatory cytokines; such an inflammatory response would negatively affect the health of the host. When compared to S.C., PAH-S.C. exhibited greater safety in all the tested parameters including bacterial accumulation in tissue, inflammatory cytokine induction and weight loss.
Because bacterial replication in tumors and the subsequent lysis of tumor cells may induce cell-mediated immune responses to tumor cells, higher oncolysis could, in part, account for increased infiltration of CD8+ T-cells in Salmonella-treated tumors.6, 21 The T-cell response against tumor cells may enhance the antitumor efficacy of Salmonella. Saccheri et al. observed an antimicrobial response present in tumors that activates cytotoxic CD8+ T-cells, which can recognize and kill tumor cells.6 The use of T-cell-deficient mice allowed us to analyze the role of T-cells in tumor-bearing mice after Salmonella administration. These results not only indicated that T-cell mechanisms are important for the control of systemic Salmonella treatment but further suggested that T-cells participate in antitumor activities directed against Salmonella. We also found that CD8+ T cells are antitumor effectors of Salmonella.
Furthermore, Salmonella induced IFN-γ production and polarized the T-cell response to a Th1-dominant state in wild-type mice but not in CD4+ T-cell-deficient mice. Thus, a bacterially activated CD4+ T-cell tumor infiltration may be a relevant source of IFN-γ in the tumor microenvironment and may contribute, a least in part, to the host antitumor immunity induced by Salmonella. The current findings and our previous reports suggest that tumor-targeted therapy involving Salmonella, which exerts tumoricidal effects and stimulates T-cell activities, represents a potential strategy for tumor treatment.30 However, persistently high levels of PAH-S.C. in tumors induced inflammatory responses, leading to the recruitment of immune cells including macrophages and neutrophils to the tumor site (Fig. 6). Hosts immunized with Salmonella can induce an antitumor response due to the ability of anti-Salmonella immune cells to kill Salmonella-infected tumor cells.31 The antitumor effects of neutrophils, in particular, after being activated by substances derived from microorganisms have also been demonstrated in various tumors.32–34
In summary, PAH shields Salmonella from neutralizing antibodies. PAH-S.C. cells display lower toxicity and improved efficacy and safety. PAH also can provide a useful platform for the chemical modification of Salmonella, perhaps allowing other chemotherapeutic drugs to bind to tumor-targeting Salmonella. By taking advantage of the tumor-targeting activity of Salmonella and the pleiotropic activities of chemotherapeutic drugs, this system appears to hold promise for tumor treatment. These results yield insight into the complex interactions between Salmonella and host immunity, maximizing the possibility of therapeutic success. Therefore, PAH-S.C. has promising potential for further clinical studies.
The authors thank L. Dvaid for carefully proof-reading the manuscript and providing valuable comments.