Bacterial Treatment of Cancer

Cancer-causing cellular agents, CT26 (for mice) and RENCA (for rabbits), were implanted. Adoptive transfer (harvesting and diagnosing tumor cells ex vivo) of splenocytes and lymphocytes was done to separate CD4+ and CD8+ cells. Computed Tomography and Positron Emission Tomography confirmed moderately immunogenic tumors in syngeneic mice and no tumors in allogeneic mice. After 24 hours, central hemorrhagic necrosis was observed in tumors followed by neutrophil chemotaxis & activation, which lasted for another 48 hours. Three to four weeks into the experiment, 34% of the initial 47 mice were completely cured of cancer while in the remaining 66% small tumor regenerated from non-necrotic areas. 10 of the previous mice were further experimented upon, out of which 8 showed immunity to even larger CT26 dosages. Two potential explanations were hypothesized for this: either CT26 cells themselves triggered a weak immune response, or their mixture with C.novyi-NT provoked an immune response. When tested further, the latter was found to be true. Adoptive transfer techniques also demonstrated that CD8+ is preventing tumor growth and not CD4+. (Agrawal et al., 2004)

C. novyi-NT was also tested on rabbits by infecting their liver with VX2 tumor model (Agrawal et al., 2004). After just four days, inflammatory infiltrates replaced tumor cells. Gas pockets, a sign of Clostridial infection, were also found in some rabbits. This suggests C.novyi-NT caused the infection. (Inflammation occurs as a response to bacterial infection. The presence of inflammation supported the presence of bacteria, and suggested that the tumors had been digested by bacteria). From a sample size of 23 rabbits, seven were completely cured. As a test for immunity, three of those rabbits were injected with larger samples of VX2 after three, five, and eleven months of the experiment, respectively. None of them developed cancer.  Apart from this, numerous other experiments emphasize the potential Clostridium (particularly C. novyi) holds for future cancer therapy (Agrawal et al., 2004).

E. Coli

E. coli is yet another bacteria used for mitigating tumor growth. This facultative anaerobe can be engineered to utilize the novel bystander killing strategy, which is instrumental in BT. In bystander killing, bacteria are induced to make a toxin that kills all the tumor cells around while keeping its own population unaffected. E. coli has been transfected with a gene that promotes the production of toxins. These toxins are responsible for killing tumor cells.

For in vivo conditions, existing vectors are not efficient enough and can only transduce less than 1% of targeted cells. It has been observed that the E. coli purine nucleoside phosphorylase (PNP) gene brings about cellular necrosis under conditions when 1 in 100 to 1 in 1000 cells express the gene product in vitro (Gadi et al., 2000).

Gadi and colleagues studied antitumor effects in the ovarian tumor model by transfecting cells with the E. coli PNP gene and further treating it with 9-(2- deoxy-erythro-pentofuranosyl)-6-methylpurine (MeP-dR) (2000). The E. coli PNP gene is controlled by a SV40 promoter. The PNP enzyme cleaves MeP-dR to liberate 6- methylpurine (MeP), a potent, membrane permeant and freely diffusible toxin. It was seen that 30 nmol MeP-dR was converted per milligram tumor cell extract per hour (or conversion units (CU)). After treatment with MeP-dR, this wipes out entire populations of tumor cells in vitro (Gadi et al., 2000).

For in vivo studies, Gadi and colleagues transfected human ovarian tumors with E. coli PNP (2000). These were excised five days after intraperitoneal (IP) implantation from the peritoneal cavities of mice in order to determine E. coli PNP enzymatic activity and the fraction of cells expressing the gene. E. coli PNP was expressed in approximately 0.1% of the tumor cells as judged by in situ hybridization. On treatment with MeP-dR, the expression of E. coli PNP at this level recorded a 30% increase in lifespan and a 49% reduction in tumor size compared to control tumors. These observations might lead one to infer that pronounced bystander killing by E. coli PNP is conferred in vivo (Gadi et al., 2000).

It is evident that vectors capable of transgene expression in as few as one in 1000 cells can produce substantial antitumor effects if the expression on a per-cell basis is very high. Therefore, minute fractions of tumor cells expressing the E. coli PNP gene can mediate measurable antitumor effects in a mouse model of human ovarian cancer (Gadi et al., 2000).

Apart from 6-methyl purine (MeP, used in the above experiment), other substrates for E. coli PNP are also available. Upon examining cytotoxicity data on various bases, Secrist III and colleagues came up with five such compounds, namely:  2-fluoroadenine (F-Ade), 6-methylpurine (MeP), 6-thioguanine, 2-amino-6- chloro-1-deazapurine, and 3-deazaguanine (1999). The first two are preferable because they are toxic to both proliferating and non-proliferating cells. The latter three act mainly on proliferating cell populations (1999). Furthermore, the nucleoside prodrug must be as non-toxic as possible to the body but a suitable substrate for E. coli PNP (Secrist III et al., 1999). All of these considerations must be kept in mind in selecting a purine analog that will convert into a toxic nucleoside metabolite when liberated into a transfected cell.

Another novel experiment, devised by Jiang and colleagues from Chonnam National University Medical School, Gwangju, demonstrates E. coli‘s integration with a pore-forming cytotoxic protein cytolysin A (ClyA) (2010). ClyA is a 34-kd protein toxin produced by E. Coli, Salmonella enterica serovar, Salmonella Typhi, and Salmonella Paratyphi A. It is transported to the surface of bacteria and is secreted without any post-translational modification. ClyA is cytotoxic toward cultured mammalian cells and induces macrophage apoptosis because of its pore-forming ability (Jiang et al., 2010). This differs from the former case, as PNP transcribed by E. coli had no toxicity of its own but required a substrate to produce toxic effects.

The setup is similar to the experiments mentioned earlier. One can engineer an E. coli strain (E. coli K-12) and use bioluminescence to detect cancer cell necrosis by specific proteins (Jiang et al., 2010).  E. coli K-12 is non-pathogenic but has been engineered to express the genes encoding ClyA to kill tumor cells. It also possesses the Lux operon to help image infected tissues (Jiang et al., 2010). The bacterial luciferase (lux) gene cassette is a flavin-dependent monooxygenase gene cassette, remarkable for its distinctive feature in transforming chemical energy to photons of visible light. Further, it can synthesize and/or scavenge all of the substrate compounds required for its production of light, conferring self-sufficiency. Its bioluminescent signal in the transformed bacteria was detected using a cooled charge-coupled device camera (Xenogen-IVIS 100; Caliper, Hopkinton, MA).

However, very few engineered E. coli cells maintained the plasmid containing ClyA over time. To overcome this, Jiang and colleagues employed a “balanced-lethal host-vector system” using the gene encoding aspartate β -semialdehyde dehydrogenase (Asd) (2010). In order to retain plasmids in vivo, the chromosomal copy of Asd in the host strain was mutated and complemented by a vector carrying both the Asd gene and the Lux gene (Asd+pLux; pALux) or the Asd gene and the ClyA gene (Asd+pClyA; pAClyA) (Jiang et al., 2010).

To evaluate the tumor growth in primary and metastatic tumor models, E. coli K-12 was first transformed with pUC19 vector containing the Lux gene (pALux) (Jiang et al., 2010). Then bioluminescent signal in the transformed bacteria was detected using a cooled charge-coupled device camera (Xenogen-IVIS 100; Caliper, Hopkinton, MA). A clone was selected and
transformed with pAClyA. Western blot analysis (done using an anti-ClyA antibody) revealed a 34-kd protein corresponding to ClyA in the bacterial pellet and the culture medium, indicating that ClyA was successfully expressed. Bacteria producing Lux and ClyA can lyse blood cells and yield a clear bioluminescent signal in areas corresponding to the hemolysis. In the experiment performed by Sheng-Nan Jiang and colleagues from Chonnam National University Medical School, Gwangju, E. coli K-12 cells carrying Lux gene and ClyA gene were injected intravenously into immunocompetent mice bearing syngeneic tumors derived from CT26 colon carcinoma cells (2010). Lux gene expression was monitored, and it was observed that bioluminescent signals were only detected in the tumors of the mice. This shows that the genetically engineered bacteria have retained their ability to target the tumor cells, and that the bacterial treatment was tumor-specific (Jiang et al., 2010).

A single treatment with E. coli expressing ClyA markedly decreased tumor growth rates initially (9 days after treatment); however, the tumors grew after that (Jiang et al., 2010). With only radiotherapy,  the tumor growth rates were retarded, but not the tumor sizes. E. Coli expressing ClyA along with radiation therapy resulted in significant shrinking of tumors and even complete disappearance of tumors in mice bearing CT26 colon cancer tumor. Furthermore, treating mice exhibiting metastatic tumor growth with E. Coli expressing ClyA markedly suppressed the tumor growth and prolonged the survival time in mice. This demonstrates the utility of genetically engineered E. coli in treating hypoxic tumor regions (Jiang et al., 2010).

Comparison and Combination Therapy

All three bacteria studied in this review possess distinct features and thus offer unique avenues of cancer treatment. E. coli can be genetically modified to release toxins that help destroy tumor cells. Clostridium deals with tumors by releasing toxins and elicits immune cells to act upon the tumors. Salmonella works by dysregulating the AKT/mTOR signaling pathway and stimulating enzymes to induce autophagy and apoptosis, respectively. Tumor cells develop microblebs (minute rounded outgrowths on the surface of cells) upon being infected by Salmonella and are destroyed as they burst and lose their viability.

Due to such enormous variations in their mechanisms, all three bacteria have some advantage over others. While Salmonella and Clostridium are lethal to cells, E. coli is made phagocytotic by genetic engineering, allowing administration via complex payloads. As facultative anaerobes, E. coli and Salmonella can thrive both in anaerobic and oxygen-rich conditions. This enables deeper tumor penetration than Clostridium – an anaerobic bacteria – which can only grow in oxygen-deficient areas of the tumor. Due to such oxygen sensitivity, its ability to eradicate tumors or tumor areas from oxygen-rich regions is severely limited. However, Clostridium is far more efficient than the other two bacteria in target specificity and evoking an immune response.

These differences become even more important if one considers using these bacteria in conjunction with other advanced techniques. These strategies fall under the umbrella term “combination therapies” (Wang et al., 2016). For example, one may use S. typhimurium to deliver herpes simplex virus thymidine kinase (HSV TK) with a beta-lactamase secretion signal to phosphorylate the prodrug ganciclovir (a nucleoside analogue) (Wang et al., 2016). Better tumor retardation and prolonged survival was observed with Salmonella/ganciclovir treatment as compared to BT alone. Combination therapy has significantly inhibited tumor growth (approximately 86.6–88.75%). Other prodrug converting enzymes such as cytosine deaminase (which converts fluorocytosine (5-FC) to 5-fluorouracil (5-FU)) have also been introduced to Salmonella vectors (Wang et al., 2016). Due to the high oxygen sensitivity of C. novyi-NT, its germination and spread are halted in well-oxygenated areas, with the bacteria forming a rim-like structure (Feng et al., 2021). Conventional chemotherapy and radiation therapy could be used to destroy the well-oxygenated cells in this rim. A combination of C. novyi-NT with these therapies might provide substantial antitumor effects (Agrawal et al., 2004).

Conclusion

Undoubtedly, administration of bacterial therapy holds promising results and can be a boon to cancer patients in the future. There are several arguments and experimental results supporting human implication of BT. It is effective against human xenografts, can develop in human wounds, and the obtained results are not species-specific (Agrawal et al., 2004).  However, numerous issues like inadequate delivery methods to the tumor sites, targeting inefficiency, limited drug production, intrinsic bacterial toxicity, and genetic instability need to be addressed.

Most payloads delivered by tumor-targeting bacteria are toxic both to normal and tumor cells, mandating precise triggering and strict control in order for BT to be safe (Duong et al., 2019). Also, newly formed blood vessels in tumors are highly disorganized. They have incomplete endothelial linings and blind loops, resulting in poor blood flow and inefficient delivery of nutrients and oxygen to regions within the tumor tissues, forming areas of anoxia (Jiang et al., 2010). These multiple hypoxic regions within the tumor make payload delivering efficiency even worse. Moreover, the ability of human tumors to evade potential immunogenic responses is a further challenge in human implication of bacterial treatment (especially the ones that function by initiating immune responses) (Agrawal et al., 2004).

Apart from these complications, a few experimental constraints also cannot be ignored. The sample sizes for tumor testing in the studies mentioned throughout this article were very small and minutely diverse, both geographically and genetically (Sieow et al., 2021). BT’s combined effects with other humanoid conditions (pathogenic deficiency diseases, genetic disorders, etc.) are yet to be studied. Thus, enormous amount of research is needed and is being done in the field of Bacterial Cancer Treatment (Sieow et al., 2021). New delivery methods, including viral or other vectors, are being developed for selective tumor targeting (Sieow et al., 2021). Advances in microbiology like improved understanding of quorum sensing and development of gene circuit modification techniques are being devised and tested every passing second (Sieow et al., 2021). Scientists have created a whole arsenal of therapeutic payloads to be delivered by engineered bacteria, each conditioned for a specific strategy (Sieow et al., 2021). In conclusion, it is pretty realistic to say that the boundaries limiting applications of BT can indeed be broken, and with further advancements, the human application of BT will surely see light of the day.