1 INTRODUCTION 1.1 Oral microbiome and its interactions with the host (oralome)

The oral cavity contains up to 1000 microbial species in total, comprised of bacteria, fungi, viral, archaea, and protozoan species that thrive in a very dynamic microenviroment.1-5 All of these microorganisms form a complex relationship among themselves, establishing a unique microbiome, known as the oral microbiome. Interestingly, the oral microbiome forms a close symbiotic relationship with human host cells in the oral cavity. Thus, the term oralome was coined to encompass not only the oral microbiome but also the host-microbial interactions that take place in the human oral cavity.5 In this sense, healthy symbiotic host-microbiome interactions between humans and these microorganisms are known as eubiosis.5-7

The microbial composition can be dramatically affected by interspecies and host-microbial interactions. These microbial changes can impact the health and disease status of the host, since eubiosis plays an essential role both in the development of natural oral physiology and host defense mechanisms.5, 8, 9 Although the oral microbiome can compensate for most overall perturbations,5, 10 some changes can profoundly affect its composition, impacting the oral commensal populations and causing an unbalanced state known as dysbiosis.5, 11

1.2 Periodontitis and oral microbiome dysbiosis

Dysbiosis is an unbalanced microbiome state that is caused by internal and/or external microbial-ecologic changes to the oral microbiome.5 This specific state has been described as capable of promoting diseases in the host.12, 13 Since periodontitis is considered an inflammatory disease that is initiated by pathogenic bacteria, the most accepted hypothesis for periodontitis initiation and progression is that there is a dysbiotic shift in the oral microbiome.5, 14 This shift is driven by an enrichment of Prevotella intermedia, Fusobacterium nucleatum, Porphyromonas gingivalis, Tannarella forsythia, and Treponema denticola species in the microbiome.14-16 Specifically, a dysbiotic oral biofilm infiltrates the gingival pocket, which then triggers the host immune response. This reaction leads to gingival tissue inflammation (gingivitis) and, ultimately, tissue degradation and periodontitis.14

Oral dysbiosis has been associated with a variety of systemic diseases and conditions, including Alzheimer's disease, diabetes, adverse pregnancy complications, and several types of cancer, including oral, gastrointestinal, lung, breast, prostate, and uterine cancer.5, 17-20 Thus, the objective of this research is to (1) evaluate the epidemiologic evidence linking periodontitis to these types of cancer, (2) provide insights into the mechanisms by which oral microbial dysbiosis can cause these cancers, and (3) summarize the evolving evidence supporting the use of probiotics and related molecules (bacteriocins) for prevention and treatment of cancer. For more details on oral host-microbial interactions and the role of oral dysbiosis on different systemic diseases, please refer to Radaic and Kapila.5

2 EVIDENCE LINKING PERIODONTITIS AND SEVERAL TYPES OF CANCER

Cancer, in a broad sense, refers to more than 277 different types of cancer diseases, each one caused by a series of genetic mutations that lead to abnormal cell proliferation and invasion.21, 22 In the United States, cancer is the second leading cause of death. It is estimated that there will be 1.8 million new cases of cancer and more than 600 000 deaths in the United States in 2020.23

Although genetic mutations are considered the main etiology agents of cancer overall, periodontitis has been recently associated with head and neck, gastrointestinal, lung, breast, prostate, and uterine cancers.18, 33 In the following sections, we will further discuss the association between periodontitis and these cancers.

2.1 Head and neck cancer

Head and neck cancer is a devastating disease, often disfiguring and debilitating affected patients. It is the sixth most common cancer worldwide, and it comprises cancers of the oral cavity, larynx, hypopharynx, and oropharynx.34, 35 In 2018, head and neck cancer accounted for approximately 706 000 new cases and 358 000 deaths worldwide.34 In the United States, head and neck cancer accounts for 3% of all cancers and approximately 65 000 Americans are diagnosed with head and neck cancer annually.23, 34, 36 Table 1 shows the estimated US incidence and the estimated new cases and deaths in the United States (for 2020)23 and worldwide (for 2018) for each head and neck cancer subtype.34 In this chapter, we will focus on oral cancer in particular.

TABLE 1. Head and neck cancer estimated incidence and estimated new cases and deaths in the United States and worldwide Head and neck cancer classification Estimated annual incidence in United States (per 100 000) Estimated new cases Estimated deaths 2020 United States23 2018 worldwide34 2020 United States23 2018 worldwide34 Oral cavity 11.7 37 35 310 354 864 7110 177 098 Larynx 3.3 37 12 370 177 422 3750 94 771 Hypopharynx <1.0 38 17 950 80 608 3640 34 984 Oropharynx — 92 887 51 005

Oral cancers have a complex etiology that includes lifestyle factors such as alcohol and tobacco usage, which are strongly associated with most head and neck cancers progression and aggressiveness.35, 39 It was recently demonstrated that alcohol consumption plus smoking have a synergetic effect in increasing head and neck cancer risk, particularly for oral and pharyngeal cancer.40 It has been demonstrated that that alcohol and tobacco-induced head and neck cancer exhibit mutations in tumor suppressor protein p53 and inactivation of the tumor suppressor p16 gene via deletion of 9p21-22.41-43

Besides tobacco and alcohol consumption, two recent cohort studies showed that poor oral hygiene decreased the survival rates of patients with oral cancers,44 whereas good oral health behaviors, such as daily tooth brushing and an annual dental visit, reduced the risk of head and neck cancer.45 These studies suggest that oral microbial dysbiosis may be an important contributor to oral cancer pathogenesis.

Recent studies have shed light on the possibility of periodontal disease–associated pathogenic bacteria having an important role in oral cancer tumorigenesis and aggressiveness. Anaerobic and facultative bacteria can colonize and grow in tumors.46-48 The possibility of pathogenic bacterial growth in tumors is currently attributed to unique pathophysiologic features known to many cancers that benefit the growth of these particular bacteria, such as impaired and abnormal vascular architecture, enhanced permeability and retention effect, low oxygen pressure/hypoxia and extensive necrosis.47 Interestingly, the main periodontal disease pathogens (ie, T. denticola, P. gingivalis, F. nucleatum, and T. forsythia) are considered facultative anaerobes and oxygen-tolerant species.19, 49-52

Particularly for oral cancer, increased salivary bacterial counts of Lactobacillus spp, Capnocytophaga gingivalis, Prevotella melaninogenica, and Streptococcus mitis and loss of Haemophilus, Neisseria, Gemella, and Aggregatibacter genera have been reported in oral cancer patients compared with normal controls.24-26, 53 Our group identified different bacterial species colonizing oral tumors compared with healthy sites and found a high fusobacterial and low streptococcal phenotype as part of the transition from primary to metastatic oral cancer.18 Interestingly, P. gingivalis and F. nucleatum (two periodontal pathogens) were detected up to 600 times more frequently in oral squamous cell carcinoma than in paracancerous and normal tissues.54, 55 Mechanistically, F. nucleatum and P. gingivalis downregulate p53 pathway56-58 and promote increased cell proliferation of tongue and oral squamous cell carcinoma up to 125 times compared with control conditions.56, 59

Dysregulation of toll-like receptor expression may also influence the host response to periodontal pathogens, which then leads to an increase in inflammation and susceptibility to periodontitis.60-62 Periodontal pathogens predominately stimulate toll-like receptor 2 and 4. This receptor activation then leads to the production of proinflammatory cytokines via regulation by transcription of nuclear factor kappa-light-chain-enhancer of activated B cells and subsequent alveolar bone resorption through the production of matrix metalloproteinases and osteoclastogenesis.61, 63-67 Nuclear factor kappa-light-chain-enhancer of activated B cells has also been identified as an integral factor in regulating various processes associated with cancer progression, such as cell survival,68 proliferation,69 and resistance to both targeted therapy and chemotherapy.70 Recently, Kamarajan et al71 demonstrated that T. denticola, P. gingivalis, and F. nucleatum enhance oral squamous cell carcinoma migration, invasion, and tumorsphere formation via integrin alpha V/focal adhesion kinase signaling; commensal bacteria were not able to trigger the same response. The authors also demonstrated that T. denticola triggers oral squamous cell carcinoma migration via crosstalk between toll-like receptor 2 and 4/myeloid differentiation primary response 88 protein and integrin alpha V/focal adhesion kinase signaling, thereby contributing to the aggressive nature of the pathogen-enhanced oral squamous cell carcinoma phenotype (Figure 1). These pathogen-mediated cancer properties were abrogated by treatment with an antimicrobial bacteriocin. Huang et al72 demonstrated that Listeria monocytogenes had a direct tumor-stimulating effect associated with its ability to activate toll-like receptor 2–dependent signaling pathways in ovary cancer cells. Moreover, the toll-like receptor 2–dependent activation of nuclear factor kappa-light-chain-enhancer of activated B cells caused by L. monocytogenes resulted in an enhanced resistance of tumor cells to chemotherapeutics. Additionally, metastasis and progression of oral tumors were essentially retarded in toll-like receptor 2 knockout mice, compared with wild-type mice.73 Thus, toll-like receptors have a tumor-stimulating effect on a variety of cancer cell types, and this mechanism may play a direct role in driving periodontal inflammation–induced carcinogenesis.

Treponema denticola drives cancer aggressiveness through toll-like receptor 2 and 4/myeloid differentiation primary response 88 protein and Integrin/focal adhesion kinase crosstalk.71 TRL2, toll-like receptor 2; MyD88, myeloid differentiation primary response 88 protein; FAK, focal adhesion kinase

Among periodontal pathogens, T. denticola has also been implicated in oropharyngeal squamous cell carcinoma, since dentilisin, a major virulence factor of T. denticola,74, 75 was found inside of the cellular cytoplasm of the majority (87%) of oropharyngeal squamous cell carcinoma tissues.76

The focus has traditionally been on bacteria when discussing microbiological aspects of oral diseases.77 However, a causal link between various microbes in human immunodeficiency virus–infected individuals has been documented.78 Specifically, inflammation is known to stimulate human immunodeficiency virus type-1 gene expression and replication, and infection by bacterial pathogens usually involves production of proinflammatory cytokines that are associated with nuclear factor kappa-light-chain-enhancer of activated B cells activation.79 Additionally, human immunodeficiency virus–infected patients show a higher incidence of squamous cell carcinoma of the oral cavity and anus.80, 81 As a means of establishing a link between the two diseases, Imai et al78 examined the effects of P. gingivalis on human immunodeficiency virus type-1 replication. The group readily demonstrated that butyric acid produced by P. gingivalis promoted increased expression of latent human immunodeficiency virus type-1–associated genes by inhibiting histone deacetylases and enriching for acetylation at histone 3, highlighting the role of bacteria as a risk factor for promoting acquired immune-deficiency syndrome progression. Similar findings in the intestine with other butyric acid–producing bacteria, such as Clostridium, Fusobacterium, and Eubacterium, suggest that these bacteria might also be involved in the accelerated replication of human immunodeficiency virus type-1.82 Latent human immunodeficiency virus type-1 proviruses also carry methylated histone H3, which has been either trimethylated on lysine 9 or lysine 2783, 84 or dimethylated on lysine 9.85 Each of these modified histones is considered to be a repressive mark for cellular genes.86 Immunohistochemical staining of diseased periodontal epithelium revealed an increased abundance of the histone lysine-specific demethylase 4B that correlates with inflammation in murine sections exposed to Aggregatibacter actinomycetemcomitans lipopolysaccharide.85, 87 Taken together, these data suggest that bacteria-virus interactions play an integral role in promoting carcinogenesis in the oral cavity.

Alcohol88 and tobacco89 exposure influence the oral microbiome, increasing the prevalence of periodontal pathogens from Fusobacterium, Cardiobacterium, Synergistes, Atopobium, Bifidobacterium, Lactobacillus, and Selenomonas genera and decreasing the levels of Capnocytophaga, Neisseria, Haemophilus, and Aggregatibacter compared with nonsmokers.89, 90 Among the enriched bacteria, F. nucleatum is also able to damage host cell deoxyribonucleic acid (DNA).56 Thus, it is possible that F. nucleatum may play a more prominent role in alcohol and tobacco-induced oral cancer, although these associations need further testing. On the other hand, a nested case-control study with 129 head and neck cancer patients demonstrated that Corynebacterium and Kingella species were associated with a strongly reduced risk of oral cancer in those with a history of tobacco use.91 Interestingly, these genera are functionally associated with xenobiotic biodegradative metabolic pathways, including the capacity to metabolize toxicants found in cigarette smoke.90

Despite numerous studies focusing on dysbiosis in oral cancer, most of these studies did not test for the human papillomavirus status of the samples.92 Among those that did, a distinct oral microbiome composition for human papillomavirus–positive oral cancer compared with human papillomavirus–negative oral cancer was revealed, namely an enrichment in Lactobacillus, Gemella, and Leuconostoc and Weeksellaceae genera in human papillomavirus–positive tumors compared with human papillomavirus–negative tumors.25, 55, 92-94 Interestingly, species from these four bacterial genera have been recently associated with the establishment and progression of oral cancer,95-98 implicating human papillomavirus as a driver of oropharyngeal and oral cancer tumorigenesis by influencing the composition of the oral microbiome.25, 92, 93

2.1.1 Mechanisms of oral microbiome dysbiosis–induced head and neck cancer

At least four main mechanisms have been proposed to explain how oral microbial dysbiosis can induce head and neck cancer carcinogenesis (Figure 2). All these mechanisms are not mutually exclusive, and they might even occur concurrently in mediating carcinogenesis.

Epithelial barrier disruption, bacterial invasion, chronic inflammation, and genetic and epigenetic modulation are mechanisms by which an oral microbiome dysbiosis can promote carcinogenesis. AMP, antimicrobial peptides

Epithelial barrier disruption

To safeguard tissue homeostasis, the oral cavity relies on an anatomic separation between the host and the microbes, known as the oral and gingival epithelial barrier, comprised of a complex immunological network (eg, continuous neutrophil recruitment and extravasation into healthy sites) and antimicrobial peptides (eg, histatins and LL-37 produced by salivary glands).99-101 Maintaining the integrity of this barrier is key to promoting healthy host-microbial interactions.98

Several studies have demonstrated that microbes can affect this epithelial barrier function.101, 102 A proposed mechanism highlights that oral microbial dysbiosis induces epithelial barrier dysfunction, leading to head and neck cancer. A key example of this mechanism has been demonstrated with mucin-2 knockout mice, wherein gastrointestinal mucosa lacking mucin spontaneously develop colorectal cancer, but antibiotic treatment or a germ-free environment significantly reduced tumorigenesis.103 From a mechanistic standpoint, several oral microorganisms—such as Bifidobacterium, Lactobacillus, Lactococcus, Leuconostoc, Pediococcus, Peptostreptococcus, and Streptococcus spp—produce compounds, such as acids (eg, lactic, acetic, butyric, and isobutyric acids), that can alter the homeostasis of epithelial barriers, thereby disrupting the mucous barrier and leading to mucosal dysfunction.104-106 Although this mechanism has not yet been confirmed in the oral cavity, this seems relevant to head and neck cancer, since oral pathogens can degrade tight junction–associated proteins that regulate epithelial barrier function. For example, P. gingivalis can degrade gingival epithelial junctional adhesion molecule 1 in gingival cells, and T. denticola can degrade zonula occludens-1, claudin-1, and occludin.107, 108 The degradation of epithelial tight junctions enables increased permeability of the gingival epithelium, allowing bacterial virulence factors to penetrate further into the tissue and leading to bacterial invasion of the tissue,107 as T. denticola, P. gingivalis, and F. nucleatum are also able to intracellularly invade epithelial and gingival tissues.109-113

Chronic inflammation

A second main mechanism is chronic inflammation. Interestingly, periodontal disease is characterized by chronic inflammation of the supporting tissues of the teeth, which can lead to loss of periodontal ligament and alveolar bone.114, 115 The disease is driven by a dysbiotic oral microbiome that interacts with the human host and leads to inflammation of the surrounding tissues.114, 116, 117 In this context, several pathogenic bacteria, such as P. gingivalis and F. nucleatum, are enriched. These pathogens are able to upregulate several cytokines and inflammatory mediators (eg, interleukins, matrix-metalloproteinases, and tumor necrosis factor alpha) in the surrounding tissues, but also facilitate invading, persisting, and spreading to adjacent cells, promoting chronic inflammation. This chronic inflammation can lead to alterations in cell metabolism, proliferation, and tumorigenesis.109, 110, 114, 118-122

Genetic damage

A third main mechanism is genetic damage. Several lactobacilli (eg, Lactobacillus acidophilus, Lactobacillus fermentum, Lactobacillus jensenii, and Lactobacillus minutus) and streptococci species (eg, Streptococcus gordonii, S. mitis, Streptococcus oligofermentans, and Streptococcus oralis) produce reactive oxygen species, nitrogen reactive species, sulfides, nitrosamines, and acetaldehyde98, 132 that can lead to DNA damage in epithelial cells and, thus, promote tumorigenesis.125-127, 131, 133 For example, in alcohol-induced oral cancer, some oral microbial species, such as S. gordonii, S. mitis, S. oralis, Streptococcus salivarius, and Candida albicans, can metabolize ethanol to acetaldehyde. Acetaldehyde is an electrophilic molecule that reacts with nucleosides, forming DNA adducts.134-136 DNA adducts are sections of the DNA strand that are covalently bound to chemical compounds, which cause abnormalities during DNA replication that can lead to genetic mutations.137

Epigenetic modulation

Epigenetic modifications are defined as heritable alterations, not coincident with alterations in the underlying DNA sequence, that allow biologic systems to interfere with transcription in response to a variety of environmental stimuli.138 Altered expression of DNA methylation138-141 and histone modifications142-146 have been reported to play critical roles in the onset and progression of both chronic periodontitis and oral squamous cell carcinoma.147 Thus, a fourth mechanism is related to epigenetic modulation of the hosts’ gene expression.

Multiple studies have suggested that the hypomethylation status of the interleukin-6 and interleukin-8 gene promoters may be related to an overexpression of these cytokines in inflamed periodontal disease tissues compared with controls.148-150 It has been found that human gingival epithelial cells may be triggered by the release of these proinflammatory mediators, which may promote the recruitment and activation of inflammatory cells to facilitate oral malignant transformation,151, 152 further supporting a link between periodontal infection and induction of cancer-related cellular processes. Although the carcinogenesis-associated epigenetic components governing interleukin-6 and interleukin-8 have yet to be fully determined, one might consider the connections between bacterial/microbial infection, cytokines, and cancer development as intriguing and intersecting points for systems biology that may reveal novel insights for the development of better diagnostics and gene and drug therapies.153

DNA methylation and histone modifications are not separate events; they are linked and result in a unique tissue and cell-specific gene expression.154 A recent study showed that the activation of toll-like receptors by periodontal pathogens not only induced activation of nuclear factor kappa-light-chain-enhancer of activated B cells but also led to an enrichment in histone acetylation in oral epithelial cells.144 Histones are subjected to a myriad of post-translational modifications, which can open and closed regions of DNA that regulate the accessibility of transcription factors to bind to their targets.155 Interestingly, P. gingivalis and F. nucleatum can induce histone modifications, such as induction of histone acetylation, in oral epithelial cells.144 On the other hand, T. denticola seems to regulate cellular division/chromatid segregation (potentially inducing genetic instability) and histone methylation and acetylation, as aurora kinase, histone methyltransferase, and histone acetyltransferases inhibition seem to modulate matrix metalloproteinase-2 activation and expression in periodontal ligament cells in the context of T. denticola infection.156 Yet, few studies are available on histone modifications in periodontitis.

Although genetic and epigenetic similarities have been reported between chronic periodontitis and head and neck cancer,152, 157 knowledge in this domain lacks deeper mechanistic insight and is largely comprised of correlative findings.

2.2 Gastrointestinal cancer

Gastrointestinal cancer, one of the most common causes of cancer worldwide, is often subdivided by anatomic location: the esophagus, stomach, liver, gallbladder, pancreas, colon, rectum, anus cancer. Combined, gastrointestinal cancers accounted for almost 5 million new cases and more than 3.5 million deaths worldwide in 2018.34 In the United States in 2020, more than 330 000 Americans are expected to be diagnosed with gastrointestinal cancer and more than 155 000 deaths from gastrointestinal cancer are expected.23, 34 Table 2 shows the estimated new cases and deaths in the United States (for 2020)23 and worldwide (for 2018) for each gastrointestinal cancer subtype.34 In this particular section, we will focus on gastric, pancreatic, and colon cancers.

TABLE 2. Gastrointestinal cancer estimated new cases and deaths in the United States and worldwide Gastrointestinal cancer classification Estimated new cases Estimated deaths 2020 United States23 2018 worldwide34 2020 United States23 2018 worldwide34 Esophagus 18 440 572 034 16 170 508 585 Stomach 27 600 1 033 701 11 010 782 685 Colon 104 610 1 096 601 53 200 551 269 Other intestines 11 110