Combining anaerobic bacterial oncolysis with vaccination that blocks interleukin-10 signaling may achieve better outcomes for late stage cancer management

ABSTRACT

Late stage solid tumors cause significant cancer mortality rates worldwide and effective therapy remains a big challenge. Cancer therapeutic vaccines elicit tumor specific T cells that kill tumor cells yet often fail to result in tumor destruction because of the limited T cell response and the local immune-suppressive environment. Blocking interleukin 10 (IL-10) signaling at the time of therapeutic vaccination elicits much stronger T cell responses than vaccination without IL-10 blocking. Anaerobic oncolytic bacteria target hypoxic regions of the late stage tumor tissues which not only stops tumor growth but also provides a pro-inflammatory environment that may increase the effectiveness of a therapeutic vaccine by recruiting more effector T cells to tumor site. In this review, we argue that combining both bacterial and vaccine therapies may improve the efficiency of late stage cancer management.

Keywords:

• anaerobic bacterial oncolysis
• interleukin 10 (IL-10)
• therapeutic vaccine

Introduction

Cancer is a class of diseases characterized by the uncontrolled division of cells which invade adjacent tissues and distant organs using a variety of mechanisms.¹ Cancer represents a major public health problem, accounts for approximately 8 million deaths each year worldwide.¹ Approximate 90% of cancers are solid tumors sharing common structures and tumor microenvironments.^2,3 Current treatments of solid tumors include surgical removal of tumor mass following early diagnosis. Chemotherapy, radiotherapy or a combination of both are used to treat late stage inoperable solid tumors but this can result in severe side effects.⁴ New therapies, such as small molecules targeting specific cancer cell signaling pathways.⁵ and gene therapies.⁵ are under intensive investigation. However, the clinical efficacy of these treatments has yet to be fully assessed.

More than 80% of solid tumors are diagnosed when surgery is no longer an option; chemotherapy and radiotherapy are the prime choices at this stage. Both therapies often have significant side effects as normal tissues and organs are indiscriminately effected.⁶ Therefore, new targeted cancer treatment strategies need to be developed as cancers and mortality rates are expected to increase.^7,8

Tumors have a complex structure consisting of cancer cells and stromal cells (i.e. fibroblasts and inflammatory cells) that are embedded in an extracellular matrix and nourished by an abnormal vascular network.^9,10 Compared to normal tissues, solid tumor stroma is associated with a changed extracellular matrix and an increased number of stromal cells that synthesize growth factors, chemokines and adhesion molecules.^9,11-16

Late stage solid tumors contain regions of hypoxia¹⁷

Solid tumors at advanced stages show angiogenesis of abnormal blood vessels restricting blood supply and resulting in reduced oxygen levels in the tumor microenvironment.^18,19 As a consequence, some areas of the late stage tumor are anaerobic and necrotic. Tumor cells in hypoxic regions adjacent to the necrotic area may be viable ²⁰ but are likely to have a decreased supply of nutrients such as glucose and essential amino acids.^21-23 Lack of blood vessels and hypoxia may also prevent therapeutic agents reaching tumor cells at desired concentrations.²⁴ Hypoxia has been demonstrated to increase the invasionand metastasis of late stage melanoma.²⁵ If tumor cells close to blood vessels are killed, the nutrient supply to previously hypoxic cells may increase, allowing cells to survive and regenerate the tumor.^21-23,26

The tumor microenvironment is immunosuppressive

The concept that the tumor microenvironment (TME) is critical for cancer development originated in the ‘seed and soil' hypothesis proposed by Paget.^27-29 The TME not only promotes the development of cancer but also prevents immune effector cells from killing tumor cells.³⁰ The TME is complex and consists of many cell types including endothelial cells and their precursors, smooth-muscle cells, fibroblasts, myofibroblasts, neutrophils, granulocytes (eosinophils and basophils), mast cells, T, B and natural killer lymphocytes, and antigen presenting cells (macrophages and dendritic cells).³¹ All the cells within the TME can participate in tumor progression and regression, depending on the stage of tumor development. It is proposed that the TME is a dynamic milieu that is in constant evolution.³² Immune and stromal cells can interact with each other to maintain the immune suppressive environment.³²

T cell infiltration in TME has been shown to have positive prognostic import.³³ A high ratio of CD8+ T cells to Foxp3+ regulatory T cells in the TMC has been suggested to have a favorable clinical outcome in ovarian and cervical cancers.^34,35 However, analysis also demonstrates that some tumor infiltrating T cells are anergic, expressing high levels of LAG-3, and secrete few cytokines,^36,37 even though effective T cells are present in circulation.³⁶

It is accepted that tumor associated dendritic cells are deficient at present tumor associated antigen (TAA) to T cells.^38,39 The presence of CD8+DCs in TME is beneficial for tumor regression but the immune-regulatory properties of plasmacytoid DCs (pDCs) in the EME have been suggested.⁴⁰ M2 type tumor infiltrating macrophages promote cancer development and hinder cancer immunotherapy.^41-43 Tumor specific regulatory T cells, which down regulate immune responses, are often enriched at tumor site.⁴⁴ Moreover, increased levels of immunosuppressive cytokines, such as interleukin 10 and TGFβ, have also been observed systematically and locally at tumor site.⁴⁵

Targeting the tumor microenvironment^38,46-50

Anaerobic oncolytic bacteria treatment targets cancer microenvironment

Patients with large inoperable tumors have been observed to become tumor-free after fever.⁵¹ Intravenous (i.v.) injected spores from an obligate anaerobic bacterium, germinated and proliferated inside the hypoxic environment of the solid tumor.⁵² These regions, that limit the effectiveness of conventional therapies, provided a suitable environment for the proliferation of the anaerobic bacteria and an opportunity for exploring a novel method for the treatment of late solid tumors.

Anaerobic and facultative anaerobic bacteria tested to date fall into 3 classes.⁵³: (i) lactic acid, Gram-positive anaerobic bacteria such as Bifidobacterium longum, Bifidobacterium infantis and Bifidobacterium adolescentis; (ii) intracellular, Gram-negative facultative anaerobes such as Salmonella typhimurium; and (iii) obligately anaerobic, Gram-positive saccharolytic/proteolytic clostridia such as Clostridium novyi. When spores of Clostridium novyi are intravenously injected into animals, they germinate exclusively within the hypoxic region of tumors.^54,55 Approximately 30% of mice treated with Clostridium novvi spores were cured of their cancers. It is recorded that small tumors (<150 mm3) are not affected by the oncolytic bacteria treatment but that very large tumors (>450 mm3) showed substantial necrosis followed by shrinkage. It appears that an optimal treatment window exists when tumors are approximately 250–300 mm3 because the efficacy of oncolytic bacteria treatment depends on the hypoxia region of the tumor, and the larger the tumor size the more extensive is the hypoxia region.⁵⁶ Similar success rates were recorded in rabbits with intrahepatic tumors and in rats with intra- cranial tumors.⁵⁷ All cured animals rejected a subsequent challenge of the same tumor, suggesting the mechanism underlying this effect is immune mediated.²⁰ Clostridium novyi spores can also germinate and proliferate within tumors of dogs.⁵⁸ Recently it has been shown in a natural occurring canine solid tumor that intra-tumor injection of C. novyni NT spores results in substantial tumor regression in 37% in dogs, with 3 instances of complete tumor regressions.⁵⁸ Administration of C. novyi NT to a human patient with advanced leiomyosarcoma resulted in tumor regression.^58,59

In contrast to other therapies, side effects of anaerobic bacteria oncolytic therapies can be controlled by using antibiotics. For example, C. novyi-NT is highly sensitive to clindamycin and various formulations of penicillin.⁶⁰

Pro-inflammatory tumor microenvironment by oncolytic cancer therapy

When bacteria grow in a tumor they may change the environment from immunosuppressive to a pro-inflammatory (Table 1 and 2). Bacterial infections are accompanied by the release of pathogen-associated molecular patterns (PAMPs) including lipopolysaccharides (LPS) from bacteria and heat shock protein (Hsp) from necrotic cells.⁶¹ LPS and Hsp.⁷⁰ induce maturation of dendritic cells (DC), the professional antigen-presenting cells that are essential for a potent immune responses. PAMPs interact with Toll-like receptors (TLRs), leading to up-regulation of co-stimulatory molecules on dendritic cells and secretion of pro-inflammatory cytokines. These, in turn, induce the production of IFN-γ by T cells and initiate a Th1-dependent cell-mediated response.^62,63 It has been shown that administration of anaerobic bacteria results in significant enhanced tumor regression in wild-type C3H/HeN mice than in TLR4-deficient C3H/HeJ mice, suggesting that Toll-like receptor 4 mediates an antitumor host response induced by anaerobic bacteria treatment.⁶⁴ Furthermore, compared with C3H/HeJ mice intra-tumor IFNγ, CXCL9 and CXCL10 levels are significantly increased in C3H/HeN mice after anaerobic treatment. More infiltration of macrophages, neutrophils, CD4+ and CD8+ T cells in C3H/HeN mice after bacteria treatment compared with those in TLR 4 deficient C3H/HeJ mice.⁶⁴

Table 1. Combing both Anaerobic bacteria oncolysis and therapeutic vaccination may achieve better outcome for late stage cancer management.

Ref	Year	Strategy/Result
^104	1947	Intravenous injection of C.histolyticium spores resulted in tumor lysis
^68	1957	Immunosurvilliance theroy formly formulated
^76	2003	Immunization and IL10 signaling blockade enhances vaccine induced CD8 T cell responses
^105	2005	Recombinant C. acetobutylicum for IL2 resulted in enhanced anti-tumor effect
^65	2008	Anarobic bacteria treatment leads to chemokine secretion and recruit Tcells to tumor site
^71	2010	First therapeutic vaccine against prostate cancer approved by FDA
^58	2012	Clostridium novyi NT spores elicits therapeutic effects in dog naturally occurring tumors
^46	2014	Immunization and IL10 signaling blockade prevent tumor growth in a mouse transplant tumor model
^59	2014	Administration of C. novyi NT to a patient with advanced leiomyosarcoma was effective at inducing tumor regression

Table 2. Bacterial oncolysis result in local tumor inflammation responses.

Bacteriolytic therapy	Inflammatory responses in tumor	Systemic inflammatory responses	Year	References
Clostridium novyi-NT	Neutrophil, Monocytes and Lymphocytes infiltration	Increased IL6, MIP-2, G-CSF, TIMP-1 CD8+ T cells are tumor killing effector cells	2004	^20
Salmonella	Expression of IFNγ, IP-10; CD4, CD8 T cells infiltration Increased cell death		2008	^65
Salmonella	Increased expression of IFNγ, MIG, IP-10	Increased IFNγ Bacteriolytic effect is TLR4 dependent	2008	^64
Salmonella	Neutrophil, Macrophage infiltration	T cell deficient mice have lower tumor inhibition rate compared with wild-type mice	2011	^66

Bacteria inside tumors destroy adjacent cancer cells through the secretion of lipases, proteases and other enzymes.³ and clostridial spores have been shown to induce strong inflammatory responses and leucocytosis.^20,64 Systemic administration of C. novyi-NT spores results in the spores being distributed throughout the body. However, they germinate only within anoxic or markedly hypoxic regions of tumors. C. novyi-NT spores and the bacteria induce the production of interleukin 6 (IL-6), macrophage inflammatory protein 2-α (MIP-2), granulocyte-colony stimulating factor (G-CSF) and metallopeptidase inhibitor 1 (TIMP-1) that attract a massive influx of inflammatory cells.²⁰ These events, initiated largely by neutrophils, are followed within days by monocyte and lymphocyte infiltration within the tumor. The accumulation of Salmonella, a Gram-negative bacterium, within the tumor site, leads to CXCL9 and CXCL10 expression,⁶⁴ which recruits T cells to the tumor.⁶⁵ Cytotoxic T cells have been observed to accumulate within the tumor treated with oncolytic bacteria.^66,67 Furthermore, oncolytic bacteria can be genetically modified to express cytokines that have anti-tumor activity.⁶⁷ which can change the tumor microenvironment from immune suppressive to pro-inflammatory.

Administration of anaerobic bacteria does not eradicate late stage tumors but the procedure reduces tumor volume and burden and generates a pro-inflammatory tumor microenvironment. It overcomes the problems with therapeutic vaccines, which often fail to show efficacy in clinic, because of the immune suppressive environment at the late stage tumor site prevents vaccine induced T cells from killing tumor cells.^10,30,31,38 The administration of anaerobic bacteria, which can reduce tumor mass and lead to a pro-inflammatory condition at the tumor site, followed by vaccine treatment may improve the efficacy of a therapeutic vaccine.

Therapeutic vaccines against cancer

In the 1950s, the immune surveillance hypothesis.⁶⁸ proposed that the immune system of the host recognizes antigens of newly arising tumors and eliminates these tumors before they become clinically evident. In contrast, the immune system can also contribute to tumor escape to form a more aggressive form of tumor cells.⁶⁸ Progressive cancer, in which the tumor cells escape the efficient control of the immune system, was considered a rare event.⁶⁸ Therefore, activating existing immune effectors, or generating new effector cells that are able to kill tumor cells, may eradicate tumors.⁶⁹ Therapeutic vaccines against cancer are still in the early stages of development and very few show efficacy in clinical trials. Only a dendritic cell based vaccine against prostate cancer was approved by the FDA in 2010.^70,71 An ideal therapeutic vaccine should produce effector T cells which migrate to the tumor sites and kill the tumor cells.^72,73 However, current therapeutic vaccines often fail in clinical trials, due to their poor ability to elicit a sufficient number of the effector CD8+ T cells critical for killing tumor cells andthe tumor immune suppressive environment.⁷⁴

Patients with cancer often develop natural humoral, CD4+ and CD8+ T cell responses to tumor antigens.⁵⁵ The ineffective immunity to tumor antigen inhibits the effectiveness of therapeutic vaccines. Pre-existing immunity to tumor antigens inhibits the effectiveness of therapeutic vaccines and the inhibition is dependent on antigen experienced CD4+ T cells.^46,75,76 Antigen-experienced IL10 secreting CD4+GITR+ T cells (GITR = glucocorticoid induced TNF receptor) may inhibit CTL responses through IFNγ signaling. IL10 secreting CD4+ T cells can be amplified after vaccination and, therefore, prevent the generation of therapeutic vaccine induced tumor killing CD8+ T cells.⁷⁷ IL10 or TGFβ have been identified as inhibiting the effectiveness of therapeutic vaccines.^70,78 As described above, the tumor microenvironment contains immune suppressive cells, such as tumor associated macrophages, dendritic cells and myeloid derived suppressor cells, as well as immune suppressive cytokines such as IL-10 and TGFβ.

To further enhance the efficacy of cancer therapeutic vaccines, strategies that have been proposed and utilized include: (i) increasing the immunogenicity of antigens, such as substitute key animo acids of an antigenic epitope, to enhance antigenicity ; (ii) increasing the vaccine induced immune response to higher quantity and quality by using a synergistic combination of cytokines, Toll like receptor ligands and co-stimulatory molecules; and (iii) removing the braking mechanisms mediated by regulatory T cells, NKT cells and suppressive molecules such as CTLA-4, PD-1, TGF β and IL10.^79,80

Interleukin 10

Interleukin (IL)-10 was first described as a cytokine synthesis inhibitory factor, produced by mouse Th2 cells and that inhibited the activation of, and cytokine production by, Th1 cells.^81,82 The human IL-10 is a homodimer with a molecular mass of 37 kDa. Each monomer has a molecular mass of 18.5 kDa and consists of 160 amino acids.⁸³ Murine and human IL-10 share 80% sequence similarity. There are several viral IL-10 homologs: Epstein-Barr virus (BCRF1),^84,85 herpes virus type 2,⁸⁶ cytomegalovirus.⁸⁷ and Orf virus.⁸⁸ The structure of human IL-10 (ebvIL-10) has been studied by X-ray crystal-topography,^89,90 IL-10 can be secreted by various cell populations including T cell subsets (Th2, Th17, Tc2, Tr1), monocytes and macrophages.⁸²

The IL-10 receptor belongs to the class II cytokine family and is composed of IL-10R1 and IL-10R2 subunits.^91,92 IL-10RI binds to IL-10 with high affinity and is expressed by most haematopoietic  cells, although generally at low levels. T cells is downregulated upon T cell activation at both the mRNA and protein levels, however, there are activated T cells that up-regulate IL-10R expression.⁸¹ IL-10R1 expression has also been observed on non-hemopoietic cells but at a much lower level. IL-10R 1 expression was induced in fibroblasts by lipopolysaccharides and in epidermal cells or keratinocytes.

IL-10R2 is the accessary unit for IL-10 signaling and is constitutively expressed in most cells and tissues. There is no evidence for significant activation-associated regulation of IL-10R2 expression in immune cells.⁸¹

IL-10/IL-10R interaction engages the tyrosine kinases Jak1 and Tyk2, which are constitutively associated with IL-10R1 and IL-10R2, respectively. IL-10 induces tyrosine phosphorylation and activation of the latent transcription factors stat3.⁹³ The main biological function of IL-10 is to suppress the inflammatory response and regulate the differentiation and proliferation of T cells, B cells, natural killer cells, antigen-presenting cells, mast cells, and granulocytes. IL10 acts on APCs by down-regulation of MHC II, co-stimulatory molecules expression and the production of reactive oxygen and nitrogen intermediates. IL-10 also acts directly on IL10 receptor expressing T cells to reduce their cytokine production and their pathological effects.^45,73,94 Numerous investigations, including expression analyses in patients and in vitro and animal experiments, suggest a major impact of IL-10 in inflammatory, malignant, and autoimmune diseases. IL-10 overexpression was found in certain tumors, such as melanomas and several lymphomas, and is considered to be a factor in promoting further tumor development.^95,96 In contrast, a relative IL-10 deficiency is regarded to be of pathophysiological relevance in certain inflammatory disorders characterized by a type 1 cytokine pattern such as psoriasis.⁹⁷

Increase of cytotoxic T cell responses by IL-10 signaling blockade plus immunization

Immunization in the presence of IL10 inhibitors increases vaccine induced cytotoxic T cell responses^{72,75,76,79,98,99} Intra-tumor injection of a Toll-like receptor 9 ligand (CpG) along with anti-IL10 receptor antibody intra-peritoneally, leads to anti-tumor therapeutic activity and induces tumor rejection in C26 and B16 tumor models.³⁸ The therapeutic effect was attributed to the improved tumor infiltrating dendritic cells (TIDCs) and macrophage.¹⁰⁰ function after the CpG treatment and IL-10 signaling blockade in vivo. CpG or anti-IL-10 receptor antibody administration alone, or lipopolysaccharide, interferon gamma and anti-CD40 antibodies administration cannot restore the function of TIDCs. Human papillomavirus (HPV) long Early protein 7 (E7) peptide/LPS plus blocking IL-10 signaling by intraperitoneally administration of anti-IL10 receptor antibodies, inhibit HPV E7 transformed TC-1 tumor growth comparable to that induced by long E7 peptide/IFA immunization.⁴⁶ In a clinical trial, Long E7 peptide/IFA immunization has been demonstrated to be effective in the treatment of HPV16+ valve neoplasia.^101-103 These results suggest that blockingIL-10 signaling at the time of immunization can prevent tumor growth.

Bacterial oncolysis combined with novel cancer therapeutic vaccine may lead to better treatment outcomes for late stage cancers

Effective cancer treatment requires the development of novel and effective therapies with minimal side effects. Bacteria oncolysis and therapeutic vaccines are emerging as therapies with great potential and combinations of both may prove to be an effective way of managing cancer, especially at the later stages where surgical removal of the tumor mass is no longer an option. Clostridial oncolysis is able to lyse late stage cancer cells thereby reducing the tumor burden and providing a pro-inflammatory environment within the tumor. Interleukin 10 blockade at the time of immunization induces stronger vaccine induced T cell responses than does immunization without IL10 inhibition. Increased numbers of vaccine induced T cells may migrate to tumor sites at higher rates and kill tumor cells more efficiently following anaerobic bacterial treatment.

Future directions

The anti-tumor actions of anaerobic bacteria treatment are complex but may be a consequence of the production of microbial extracellular enzymes that destroy the hypoxic cancer cells. Subsequently, immune mechanisms may participate in tumor destruction. More T cells are often recruited to the tumor site after bacterial treatment and immune competent mice have a higher tumor cure rate than immune deficient mice. Tumor bearing mice treated with anaerobic bacteria can reject subsequent tumor challenge. It is, therefore, important to investigate how anaerobic bacterial treatment changes the immune suppressive environment of the tumor. It is uncertain as to whether tumor associated dendritic cells or macrophages within the tumor site become immunogenic. M2 type macrophages become M1 type, N2 type neutrophils become N1 type neutrophils. It is not known if bacterial treatment reduces the immune suppressive cytokine levels especially interleukin 10 or TGFβ levels at the tumor site. In addition, the way that T cells are recruited to the tumor side after bacterial oncolysis awaits clarification, as does whether tumor cells are more sensitive to killer cells after bacterial treatment. Once the mechanisms of bacterial oncolysis are understood investigations should focus on whether combining both bacterial oncolysis with a novel therapeutic vaccine incorporating IL10 signaling inhibitor better improves the outcome of cancer management in animal models prior to clinical trials.

Disclosure of potential conflicts of interest

The authors do not have potential conflicts of interest.

Acknowledgments

The authors wish to thank Prof. Richard Burns for critical reading of the manuscript.

Funding

The current research was supported in part by Science and Technology Research program of Foshan city (No.: 2012B03180003); Science and technology research program of Guangdong province (No.: 2012AA100461) to XSL and National Natural Science Foundation of China (No.: 81472451) to XSL and TFW.