Cancer, a diagnosis that still hits patients very hard. Despite intensive research as well as newly developed improved therapies, cancer remains the second most frequent cause of death in industrialized countries. Increased life expectancy is certainly one reason for this since cancer is essentially a disease of old age. To date, cancer is mostly treated by surgery, radiotherapy or chemotherapy. However, not every cancerous tissue can be targeted with a scalpel and physical and chemical methods do in general not distinguish between healthy and malignant tissue. For such reasons, cancer research is presently strongly investigating biological treatments like immunotherapy which are supposedly more specific.
The History of Bacteria as Cancer Therapeutics
The idea to use bacteria as cancer therapeutic is not new. It only had to be revitalized. As early as at the beginning of the 19th century, scientists had noticed a correlation between bacterial infection and tumor regression. In 1868, the first ‘clinical trial’ for bacteria mediated cancer therapy was performed by W. Busch in Berlin. He purposely infected a female cancer patient by transferring her into a contaminated bed in which a patient had died of Erysipelas (infection by Streptoccocus pyogenes). As expected the patient got infected and the tumor regressed. However, the cancer patient died of the infection. At that time it was not possible to control the bacteria. Inspired by this trial, the American physician William Coley used bacteria systematically for cancer therapy. He therefore became the first pioneer of immunotherapy. After some fatal trials with live bacteria, he noticed that a balance between infection control and therapeutic benefit had to be found. As a consequence, he used a mixture of heat-inactivated S. pyogenes and Serratia marcescens known as ‘Coley’s Toxin’. Many patients with inoperable sarcomas were effectively treated with this toxin. Although Coley was quite successful, his work was not acknowledged at that time and forgotten. A possible reason for this development, in addition to upcoming radiotherapy, might have been the inability to explain the therapeutic mechanism and to control bacterial infections or side effects of the therapy.
Nevertheless, bacteria made it to the clinic as routine cancer treatment.
Since the late 80s, oncologists successfully use BCG (Bacille Calmette-Guerin, a vaccine variant of Mycobacteria bovis) to prevent relapses of bladder cancer after surgically removal of the primary tumor . Although the exact mode of action of the bacteria is not fully understood, they might enhance the immune response against the cancer cells by e.g. activation of natural killer cells.
Also the idea to use bacteria systemically i.e. intravenously for cancer therapy was undergoing a renaissance. In various animal models, it had been shown that several bacteria are selectively invading solid tumors and colonize them at high proportions. Although the mechanism of tumor colonization and thus tumor specificity is only partially understood, the induction of a cytokine storm might be important . Cytokines like Tumor Necrosis Factor-α (TNF-α) were considered to open the irregular vasculature of cancerous tissue and therefore promote bacterial infiltration (fig. 1). The resulting hemorrhage that is also macroscopically visible (fig. 2) is leading to a large necrotic region within the tumor characterized by hypoxia and high nutrient availability. Such parameters provide perfect growth conditions for facultative anaerobic bacteria like Salmonella Typhimurium. At the same time, the immune system cannot function properly under such conditions. This might explain the enrichment of bacteria in the tumor i.e. the tumor specificity of some bacteria.
Which Bacteria are Effective?
At the moment, most of the research using bacteria as cancer therapeutics is focusing on S. Typhimurium and Clostridium novyi NT . C. novyi is a gram-positive, spore forming, obligate anaerobic bacterium. Since the spores can only germinate in the absence of oxygen, this bacterium should be specific for necrotic areas in tumors that provide anaerobic conditions. Accordingly, C. novyi cannot infect oxygenated healthy tissue. To increase its safety profile, the gene encoding the α-Toxin was deleted, in addition. Aside from mouse experiments, C. novyi was already tested in pre-clinical and clinic trials with dogs and humans. Results were to some extent promising. In addition, in a recently published report, orthotopic glioblastomas were successfully targeted by C. novyi spores upon intravenous infection in a rat model . These data indicate that the spores are able to pass the blood brain barrier under such conditions. However, the restriction to anaerobic regions is also the big disadvantage of Clostridia. Metastases or small tumors without necrotic areas cannot be targeted by this therapeutic approach.
Application of Salmonella Typhimurium as Anti-Cancer Agent
We focus on S. Typhimurium. These bacteria are gram-negative, facultative anaerobic and can colonize oxygen rich as well as oxygen deprived areas (fig. 3A). Consequently, they are able not only to infect tumors but also healthy organs like liver or spleen. In order to guarantee a safe application to cancer patients, the Salmonella need to be adapted. In order to optimize Salmonella for therapy, we have decided to modify their outer membrane component Lipopolysaccharide (LPS) for two reasons: a) LPS is highly immune-stimulatory and could induce sepsis. b) Strains bearing a modified LPS structure are more susceptible to immune effector mechanisms. Therefore, Salmonella bearing a mutant LPS were generated by deleting the genes for rfaD, rfaG or rfaL and tested in our murine tumor model (fig. 3B). The shorter the LPS structure was, the safer the mutant became. Unfortunately, they were also losing their therapeutic strength. Thus, ∆rfaD was safe but ineffective while ∆rfaL was effective but harmful to the mice. However, the safety profile as well as the tumor specificity of the ∆rfaD mutant was extremely high i.e. healthy tissues were free of Salmonella two days after infection. Apparently, the strain needed to be adjusted more sophisticatedly to find a proper balance between therapeutic benefit and safety. Therefore, we placed the rfaD gene under control of a promoter that can be regulated by addition of the sugar arabinose. In the presence of arabinose, our Salmonella strain is now highly virulent since it bears a wild-type LPS structure. However, upon injection into a mammalian host, which is essentially arabinose free, the expression of rfaD is shut down. The mutant becomes strongly attenuated in vivo. Applying this strategy was very successful. By administering this initially highly virulent, wild-type-like Salmonella variant, the anti-tumor immune response was apparently very strongly enhanced. Most of the tumors were completely rejected while no mice succumb to the infection.
Obviously, we have found a good balance between therapeutic benefit and safe administration, although it still needs further optimization. Already Coley had noticed that this balance is crucial for bacteria mediated cancer therapy. Bacteria in the wild-type state efficiently attack cancerous tissue but, as Coley observed for S. pyogenes, can kill the patient. At that time, when genetic engineering was not yet known, Coley solved the problem by heat inactivation although he noticed a loss in therapeutic efficacy. It still appears to be nowadays a major obstacle in designing bacteria for therapy. An over-attenuated bacterium might be safe but cannot induce a proper anti-tumor response anymore. This might be the explanation for the negative outcome of the clinical trials using a highly attenuated Salmonella variant. We believe that a strategy combining increased immune-stimulatory capacity with adequate attenuations, like in the case of our new rfaD mutant, might be a promising strategy to tailor bacteria for cancer therapy.
Nevertheless, a therapy that only relies on the intrinsic anti-tumor effect of bacteria cannot be successful for all types of cancer. Although some tumors may be rejected, many tumors might exhibit escape mechanisms to overcome the tumor specific response induced by the bacteria. Therefore, bacteria that exhibit tumor specificity should be used as vehicles to shuttle therapeutic compounds directly into the cancerous tissue. The challenge is now to find proper bacterial secretion systems for such heterologous compounds since most of the bacterial secretion machineries are highly specific for species specific molecules. Although the distance to clinical trials is still far, we think we did a step in the right direction with our concept.
Sara Leschner, Kathrin Westphal, Nicole Dietrich, Nuno Viegas, Jadwiga Jablonska, Marcin Lyszkiewicz, Stefan Lienenklaus, Wener Frank, Nelson Gekara, Holger Loessner, Siegried Weiss: Tumor Invasion of Salmonella enterica Serovar Typhimurium Is Accompanied by Strong Hemorrhage Promoted by TNF-α PLoS one. 4(8): e6692 (2009), DOI: 10.1371/journal.pone.0006692
Verena Staedtke, Ren-Yuan Bai, Weiyun Sun, Judy Huang, Kathleen Kazuko Kibler, Betty M. Tyler, Gary L. Gallia, Kenneth Kinzler, Bert Vogelstein, Shibin Zhou, Gregory J. Riggins: Clostridium novyi-NT can cause Regression of Orthotopically Implanted Glioblastomas in Rats, Oncotarget. 6(8): 5536-5546 (2015)
, DOI: 10.18632/oncotarget.3627
Sebastian Felgner1, Dino Kocijancic1, Siegfried Weiss1,2
1 Molecular Immunology, Helmholtz Centre for Infection Research, Braunschweig, Germany
2 Institute of Immunology, Medical School, Hannover, Germany
Helmholtz Centre for Infection Research