Dynamic impact of virome on colitis and colorectal cancer: immunity, inflammation, prevention and treatment

Abstract

The gut microbiome includes a series of microorganism genomes, such as bacteriome, virome, mycobiome, etc. The gut microbiota is critically involved in intestine immunity and diseases, including inflammatory bowel disease (IBD) and colorectal cancer (CRC); however, the underlying mechanism remains incompletely understood. Clarifying the relationship between microbiota and inflammation may profoundly improve our understanding of etiology, disease progression, patient management, and the development of prevention and treatment. In this review, we discuss the latest studies of the influence of enteric viruses (i.e., commensal viruses, pathogenic viruses, and bacteriophages) in the initiation, progression, and complication of colitis and colorectal cancer, and their potential for novel preventative approaches and therapeutic application. We explore the interplay between gut viruses and host immune systems for its effects on the severity of inflammatory diseases and cancer, including both direct and indirect interactions between enteric viruses with other microbes and microbial products. Furthermore, the underlying mechanisms of the virome’s roles in gut inflammatory response have been explained to infer potential therapeutic targets with examples in specific clinical trials. Given that very limited literature has thus far discussed these various topics with the gut virome, we believe these extensive analyses may provide insight into the understanding of the molecular pathogenesis of IBD and CRC, which could help add the design of improved therapies for these important human diseases.

1. Introduction

According to GLOBOCAN 2020 released by the International Agency for Research on Cancer (IARC), colorectal cancer (CRC) is the third most common cancer in 2020 (10% incidence rate) and the second leading cause of cancer death (9.4% mortality) in the world [1]. Previous studies have shown that chronic inflammation is related to the etiology and pathogenesis of CRC, termed colitis-associated CRC (CAC) [2]. Inflammatory bowel disease (IBD), including Crohn’s disease (CD) and ulcerative colitis (UC), dramatically increases the risk of CAC [3, 4]. For instance, UC patients have a 2.4-fold increased risk of CAC occurrence [5]. CD patients have similar CAC incidence but at a younger age than the healthy population [6]. Although the molecular mechanisms of CRC remain largely unknown, one possibility is related to an aggravated immune response of genetically susceptible individuals to the gut microbiota [7, 8]. In return, the microbiota may also markedly impact the host defense and inflammatory response to shape IBD and CRC development and progression.

The gut microbiota represents a complex ecosystem and plays a vital role in human health and the pathogenesis of multiple diseases. In recent decades, trillions of bacteria, fungi, viruses, parasites, and archaea have been identified with advances in high-throughput, next-generation sequencing technology and bioinformatics. Bacteria (Helicobacter pylori for gastric cancers [9]) and viruses (hepatitis B and C viruses for liver cancers and human papillomavirus for cervical cancer [10, 11]) have been found to strongly affect cancer genesis (tumorigenesis), progression, and treatment. Colitis and cancer microbiome studies have focused almost exclusively on bacterial members’ genomes [12, 13]. At present, only 11 out of the estimated 10¹² distinct microbial species are considered human carcinogens [14, 15]. Increasing evidence suggests that the existence and quantity alterations of certain microbes alone are not enough to cause cancer, but they can serve as the second hit to promote inflammation and carcinogenesis caused by pathogenic bacterial infections, dysbiosis, and genetic defects in host immune regulation [16]. Large-scale clinical trials are currently testing the impact of the microbiota on cancer care, including diagnosis, dietary modification, and intratumoral injection of engineered bacteria [17].

Decades of searches have identified only a few viruses that can directly cause cancer, but many seem to have a complicated impact on the host’s immune system. The involvement of viral infection in IBD or CRC is increasingly being recognized [18–20]. CRC progresses in a stepwise process. Disturbances of viral homeostasis may trigger or facilitate inflammatory diseases (such as IBD), promote dysplasia, and ultimately lead to cancers with severe symptoms and high mortality [18, 19, 21]. It has been shown that viruses are directly involved in inflammation and tumorigenesis because of their ability to infect human cells and mutate [19, 22, 23]. In addition, the virus could act indirectly by regulating the stability and composition of bacterial communities [24, 25]. Thus, viruses are presumed to be potential modular biotherapeutics. Therefore, further understanding of how viruses impact IBD and CRC has great potential for early detection and prevention of progression beyond early cancer stages and significantly facilitates therapeutics development.

Here, we review aspects of the current essential knowledge that link the virus communities to inflammatory responses as well as the initiation and progression of colitis and CRC. We summarize the composition and direct function of mucosal virus in health and disease (e.g., colitis and CRC). Moreover, we discuss the viruses’ indirect role in impacting colitis and cancer by modulating the associated bacterial community. Finally, we try to fill the gaps in knowledge and attempt to point out potential future directions.

2. Distribution and abundance of viruses in the human gut

The human gut virome contains genomes of all DNA and RNA viruses naturally present in the intestine, including bacteriophages, eukaryotic viruses, archaeal viruses, and endogenous retroviruses (Figure 1). Previous research mainly focused on bacterial (bacteriophages) and eukaryotic viruses. Apart from the above, other components of the gut virome have not been studied in-depth due to limited understanding of their role in humans and a relatively small virus database.

Figure 1.

Virus distribution in the human gut. The human virome contains DNA and RNA types. Human feces contain 10⁹-10¹⁰ virus-like particles (VLPs) per gram gross weight. The main components of virome are eukaryotic, bacterial (bacteriophage, or phage), archaeal viruses, and endogenous retroviruses. The core phage community, including crass-like phage, is the most common human gut phage. The illustration was created by adapting SMART (https://smart.servier.com) and Vecteezy (vecteezy.com) templates.

Bacteriophages (phages) are the most abundant type of bacterial viruses, which can influence homeostasis through their immunomodulatory and bactericidal effect against bacterial pathogens living in the gut [26]. The human gut virome, isolated from a single healthy individual [27], was first published in 2003 with at least 10⁹-10¹⁰ virus-like particles (VLPs) per gram (Figure 1). The gut virome includes almost 10¹⁵ phages which are 10 times more than bacterial cells and 100 times more than human cells [28, 29]. The commensal viruses dominated by double-stranded DNA (dsDNA) phages Caudovirales and single-stranded DNA (ssDNA) phages Microviridae [30] are highly specific, diverse, and stable. However, the majority of these phages remain unclassified [31]. The core phage community, including crAss-like phage, is the most abundant and widespread virus in the human gut (20–50% of individuals) [32, 33]. The composition of crAss-like phage in Western society shows a dramatic difference compared to African populations [34]. In addition, a distinct prevalent phage taxon (LAK phages) has recently been described in individuals from the region of Laksam in Bangladesh and Tanzania [35]. Since the commensal viruses are involved in host immunity education and maturation [36], a dysbiosis of these viruses might lead to IBD [24, 37] and cancer [38, 39], as we will discuss later in this review.

3. Viruses that are strongly related to intestinal inflammation and cancer

A variety of enteric viruses might infect humans, including retroviruses, noroviruses, rotaviruses, adenoviruses, and herpesviruses [28]. For example, norovirus is the leading cause of food-borne gastroenteritis worldwide [40]. Murine norovirus (MNoV) establishes lifelong intestinal infection in mice, and asymptomatic individuals may carry noroviruses for a long time without apparent illness [41]. Enteric viral infection may lead to pathophysiological processes, ranging from asymptomatic to moderate and severe acute or severe chronic disease, possibly resulting in colitis and cancer (Figure 2).

Figure 2.

Host-virome interactions in intestinal inflammation and cancer. The enteric virome richness increases with the severity of IBD and CRC patients, while bacterial diversity and richness decrease, reflecting a reverse correlation of disease with the bacterial and viral microbiome. Induction of intestinal inflammation or direct genotoxic activity is the main mechanism underlying microbiome-induced intestinal carcinogenesis. Viruses can promote CRC in the following ways: ① directly infect the intestine, ② indirectly affect the host through bacteria, and ③ interact with the host’s immune system and induce immune responses. The illustration was created by adapting SMART (https://smart.servier.com) and Vecteezy (vecteezy.com) templates.

Over the last two decades, along with advances in viral metagenomics, the arena of research between virus and intestinal inflammation and cancer has progressed from detection of the presence of viral particles to the interaction and the virus-driven molecular mechanisms (Table 1).

Table 1.

Association of viruses with intestinal inflammation and cancer

Author, Year	Virus	Disease	Sample origin	Sample size	Biospecimen type	Observations	Ref
Norman et al, 2015	Caudovirales	CD, UC	UK, USA	72	Feces	CD and UC patients exhibit a significant expansion of Caudovirales compared with Microviridae.	[19]
Wagner et al, 2013	Caudovirales	CD	Australia	21	Biopsy specimens, gut wash samples	Increased abundance of Caudovirales in ileum biopsies and gut wash samples of pediatric CD patients.	[22]
Wang et al, 2015	Herpesviridae	CD, UC	Canada	10	Biopsy specimens	Increased expression of human endogenous viral sequences in patients with herpesviridae sequences.	[42]
Ungaro et al, 2019	Hepadnaviridae, Hepeviridae	CD, UC	USA	359	Biopsy specimens	Higher levels of eukaryotic Hepadnaviridae transcripts in UC patients; increased abundance of Hepeviridae in CD patients.	[43]
Cornuault et al, 2018	Faecalibacterium prausnitzii phage	CD, UC	UK, USA	171	Feces	F. prausnitzii phages are more prevalent or abundant in the fecal samples of IBD patients compared to healthy controls.	[44]
Damin et al, 2007	HPV	CRC	Brazil	72	Biopsy specimens	HPV is present in the rectum and colon of most CRC patients.	[47]
Burnett-Hartman et al, 2013	HPV	CRC	USA	555	Biopsy specimens	Very low prevalence of HPV-16 and −18 among CRC samples.	[49]
Gornick et al, 2010	HPV	CRC	USA, Israel, Spain	309	Biopsy specimens	No association of HPV with CRC.	[50]
Chen et al, 2012	CMV	CRC	Taiwan (China)	163	Biopsy specimens	Human CMV preferentially infects CRC lesions compared to normal healthy tissue.	[52]
Dimberg et al, 2013	CMV	CRC	Sweden, Vietnam	202	Biopsy specimens	Human CMV DNA rate was significantly higher in cancerous tissue compared to paired normal tissue.	[53]
Chen et al, 2014	CMV	CRC	Taiwan (China)	95	Biopsy specimens	Human CMV is associated with a worse outcome of elderly CRC patients.	[54]
Chen et al, 2016	CMV	CRC	Taiwan (China)	89	Biopsy specimens	Human CMV may influence the outcome of CRC in an age-dependent manner.	[55]
Chen et al, 2015	CMV	CRC	USA et al1	230	Biopsy specimens	Human CMV genetic polymorphisms in CRC tumor tissue are associated with different clinical outcomes.	[56]
Harkins et al, 2002	CMV	CRC	USA	29	Biopsy specimens	Association between human CMV and colon neoplasia.	[59]
Jung et al, 2008	JCV	CRC	USA	74	Biopsy specimens	JCV T-Antigen is specifically expressed in the early stage of CRC.	[61]
Shavaleh et al, 2020	JCV	CRC	Tunisia et al2	2,576	Biopsy specimens	As an oncogene virus, JCV could increase the odds of CRC.	[62]
Campello et al, 2010	JCV	CRC	Italy	185	Biopsy specimens, blood and urine samples	JCV is not detected either in normal or pathological tissues.	[63]
Gock et al, 2020	JCV	CRC	Germany	60	Biopsy specimens, gDNA, cDNA	No direct correlation between tumorigenesis and viral load.	[64]
Goel et al. 2006	JCV	CRC	USA	100	Biopsy specimens	Significant associations between JCV T-antigen and chromosomal instability in CRC.	[65]
Coelho et al, 2013	JCV	CRC	Portugal	200	Biopsy specimens	JCV may be associated with tumor suppressor genes (i.e., p53) polymorphisms.	[66]
Enam et al, 2002	JCV	CRC	USA, Mexico	27	Biopsy specimens	Possible association of JCV with CRC.	[67]
Destri et al, 2013	HBV, HCV	CRC	Italy	488	Biopsy specimens	Hepatitis virus-infected CRC patients had better 5-year disease-free survival and a lower incidence of metachronous liver metastases.	[68]
Su et al, 2020	HBV	CRC	Taiwan (China)	69,478	Biopsy specimens	Chronic HBV infection is strongly associated with increased risk for CRC.	[69]
García-Alonso et al, 2016	HCV	CRC	Spain	570	Biopsy specimens	No statistically significant differences between HCV prevalence and risk factors for CRC.	[70]

CMV, cytomegalovirus; CD, Crohn’s disease; cDNA, complementary DNA; CRC, colorectal cancer; gDNA, genomic DNA; HBV, hepatitis B virus; HCV, hepatitis C virus; HPV: human papillomavirus; JCV, John Cunningham virus; Ref, reference; UC, ulcerative colitis.

¹USA, France, Italy, Japan, China, Taiwan (China).

²Tunisia, China, Taiwan (China), Israel, USA, Iran, Portugal, Japan, Jordan, Netherlands, Italy, Greece, Poland.

3.1. Inflammatory Bowel Disease and viruses

The composition of phages from individuals with IBD is significantly different from healthy controls [19, 37]. Patients with CD or UC exhibit a significant expansion of Caudovirales compared to Microviridae and a decrease in bacterial richness and diversity [19]. Caudovirales was also found to be increased in ileal biopsies of pediatric CD patients, but that in stool was not associated with UC onset [22]. In another study, patients with Herpesviridae sequences in the colons showed increased expression of human endogenous viral genes and increased microbiome diversity [42]. Within early-diagnosed treatment-naive IBD patients, patients with UC had a higher amount of Hepadnaviridae transcripts than patients with CD and controls, with a lower concentration of Polydnaviridae and Tymoviridae. In addition, CD patients showed an increased abundance of Hepeviridae compared to controls, with a reduced abundance of Virgaviridae [43]. Faecalibacterium prausnitzii is a species that is generally depleted in IBD patients. F. prausnitzii phages were found significantly more prevalent in samples from IBD patients than those of household controls, suggesting enhanced phage-mediated mortality of F. prausnitzii in IBD [44]. Significantly, specific viral infections can interact with IBD risk genes to alter intestinal disease in IL-10- or Atg16L1-deficient mice, indicating that certain species of the virome may contribute to IBD [45, 46] (Figure 3).

Figure 3.

Potential impact of enteric viruses on intestinal epithelial cells and host immune cells. In IL-10-deficient mice, MNoV ① crosses the epithelial barrier through microfold cells, ② infects immune cells including lymphocytes, macrophages, and dendritic cells, and ③ causes an interferon response (IFN-α, β). In IL-10-deficient mice, MNoV can ④ infect tuft cells, and ⑤ induce IFN-λ secretion, which plays a critical role in regulating persistent virus in vivo. ⑥ Persistent MNoV can also drive Paneth cell abnormalities related to IBD. ⑦ Mice with Atg16L1 mutations infected by MNoV trigger the abnormality of Paneth cell and display intestinal disease. IBD, inflammatory bowel disease; IEC, intestinal epithelial cell; MNoV, murine norovirus. The illustration was created by adapting SMART (https://smart.servier.com) and Vecteezy (vecteezy.com) templates.

3.2. Colorectal cancer and viruses

There are no significant differences in viral richness between CRC patients and healthy controls [43]. However, viral dysbiosis is reportedly associated with early- and late-stage CRCs [39]. Although the influence of viruses in the pathogenesis of CRC is growing, there has been no clear and consistent consensus conclusion yet. Further rigorous experimentation and cross-cohort validation are key to solve the myth, establishing or disproving the link of viruses to CRC, especially clinical prevalence and pathophysiology.

The human papillomavirus (HPV) genome was detected in CRC specimens, indicating a possible association of HPV presence with an increased risk of developing CRC [47, 48]. However, other studies questioned the role of HPV in CRC carcinogenesis because little or no HPV DNA was found in CRC specimens [49, 50], highlighting the need for additional research to clarify such inconsistency. Compared with HPV-negative tissues, four differentially expressed genes (WNT-5A, c-Myc, MMP-7, and AXIN2) were found to be up-regulated in HPV-positive CRC samples [51], consistent with an earlier report implicated HPV’s relevance to CRC pathogenesis [23].

It has been shown that human cytomegalovirus (CMV) preferentially infects CRC lesions rather than normal healthy tissue [52, 53], which may be related to the poor prognosis of CRC patients [54–56]. The phenomenon may be involved in CRC cell proliferation and progression, with higher expression of TLR2, TLR4, NF-κB, and TNF-α in CMV-infected CRC samples compared to control tissues [57], and increased expression levels of Bcl-2, cox-2, and Wnt/β-catenin in cancer cell lines [58, 59].

The members of the Polyomaviridae family, mainly human polyomavirus 2 (known as John Cunningham virus, JCV), are found to be linked to CRC, pointing towards a possible carcinogenic role of JCV [60–62]. However, little or no JCV DNA was detected in CRC samples in some studies [63, 64]. It has been proposed that JCV promotes colon carcinogenesis in the following ways: first, an early JCV protein T-antigen (T-ag) is believed to mediate the oncogenic potential of the virus and link to chromosomal instability [63, 65]; next, JCV may be responsible for the induction of polymorphisms and/or alterations in tumor suppressor genes (i.e., p53) [66]; last, JCV may alter cell behavior, such as migration and invasion, underlining a possible involvement of PI3K/AKT, MAPK and/or Wnt/β-catenin pathways [67].

In a rare study with Italian populations, CRC patients with hepatitis B and C virus-related liver diseases exhibited better 5-year disease-free survival and a lower incidence of metachronous liver metastases, claiming a “metalloproteinase inhibitor” hypothesis rather than a direct effect of the viral infection by the authors [68]. In a separate study, hepatitis B virus (HBV) infection was closely associated with an increased risk for CRC [69]. Another study found no correlation between hepatitis C virus (HCV) infection and CRC development [70]. Due to the significant variability, it is necessary to further investigate the interrelationship between different types of hepatitis viruses, inflammatory diseases, and CRC.

3.3. The paradoxical role of phages in IBD and CRC

The same phages may act as a double-edged sword within the inflammatory environment of the gut. Caudovirales phages significantly reduce colonization by carcinogenic bacteria and increase the survival of animals predisposed to CRC development. However, the expansion of phages worsened IBD pathogenesis in mice [24]. Similarly, individuals with CD display a significant increase in the abundance of phages within the order Caudovirales [19]. It is necessary to analyze these complex interactions to test whether the viral microbiota may have therapeutic benefits in future work.

Thus, although it is debatable, studies in this area have found that some eukaryotic viruses can infect human cells, establish infections, trigger immune responses and, sometimes, cause serious diseases. Future studies should explore the alterations and impact of viral composition on specific intestinal disorders, including IBD and CRC.

4. Viruses that are likely associated with intestinal inflammation and cancer

Enteric viruses are first exposed to bacteria to initiate the replication in host cells (enterocytes or immune cells). Animal infection models have been commonly used to examine interactions between intestinal viruses and other microbes [71]. Currently, the experiments have been performed by using antibiotic-treated and germ-free mice or immune-deficient and/or young mice infected with human and murine viruses [28]. The beneficial effects of the bacterial microbiota on enteric viruses were not recognized until recently, possibly because many studies involved intraperitoneal injection viruses rather than the natural oral route [72, 73].

Since phages and bacteria are engaged in an intense arms race during evolution, phages may alter the bacterial microbiome and play a role in intestinal physiology and disease through complex mechanisms, which require extensive further elucidation (Figure 4A). First, enteric phages are responsible for the horizontal gene transfer (HGT) among bacterial communities, including pathogenesis and antibiotic resistance, which impose a healthcare burden in controlling bacterial infection [74]. Second, the activation of phages leads to the lysis of their bacterial hosts and changes in the abundance of specific gut bacterial species [75]. Last (but not least), lysis of bacteria would release proteins, lipids, and nucleic acids that serve as pathogen-associated molecular patterns (PAMP) and antigens, which may trigger inflammatory signaling cascades to induce cytokines, cellular infiltration, and tissue damage [19].

Figure 4.

Transkingdom interactions and mechanisms. A, Phage directly or indirectly interacts with the host through the host-associated bacteria. These interactions may influence host genetic variations and pose distinct effects on host health. B, The mechanisms by which bacteria enhance enteric virus replication and transmission. C, Possible mechanisms of viruses affect the intestine. ① Virus can be sensed by the dendritic cell via RIG-I–MAVS–IRF1 pathway, which stimulates IL-15 secretion, thereby enhancing the proliferation and inhibiting the apoptosis of IELs. ② Resident virus is recognized by TLR3 and TLR7 on dendritic cells, producing protective IFN-β to suppress intestinal inflammation. ③ Phage is endocytosed in dendritic cells, activating B and T cells and stimulating IFN-γ-mediated TLR9-dependent immune responses, exacerbating colitis. IEC, intestinal epithelial cell; LPS, lipopolysaccharide. The illustration was created by adapting SMART (https://smart.servier.com) and Vecteezy (vecteezy.com) templates.

Recent studies indicate that viruses would gain enhanced replication and transmission by bacteria via the following mechanisms (Figure 4B): (1) Virion stabilization by bacteria. Poliovirus (PV), spread through the fecal-oral route, could disseminate to the central nervous system. Intestinal bacteria stabilize virions and limit thermal inactivation to enhance PV replication and fecal-oral transmission in mice [76]. (2) Bacteria may increase host cell attachment. PV binds with lipopolysaccharide (LPS, a glycan on the surface of Gram-negative bacteria) and peptidoglycan (a major component of Gram-positive bacterial cell wall), which enhance the viral attachment to host cells [73, 76]. (3) Immune tolerance can be induced by virion-bound LPS. Mouse mammary tumor virus (MMTV), a member of the Retroviridae, is spread from mothers to offspring through milk in humans. MMTV binds to LPS, which induces host TLR signaling and IL-10-mediated immune tolerance and initiates viral replication and transmission [77]. (4) Host IFN-λ is likely regulated by microbiota. MNoV, a member of the norovirus genus within the Caliciviridae, is spread through the fecal-oral route. Bacteria promote MNoV replication, likely through the regulation of IFN-λ responses.

In addition to indirectly affecting the host through bacteria, phages can also directly interact with the host’s immune system and trigger immune responses. Accumulating evidence indicates that phages exert interactions with host intracellular immune pathways and activate immune responses in the intestine of individuals with IBD or CRC [24, 78, 79].

Commensal viruses have been shown to protect host animals against dextran sodium sulfate(DSS)-induced colitis via a beneficial effect on intraepithelial lymphocytes, through the RIG-I-MAVS-IRF1-IL15 axis [25], and suppression of the TLR3 and/or TLR7–IFN-β pathways [80] (Figure 4C). Caudovirales phages stimulated an immune response and aggravated colitis indirectly and directly. First, phages lyse bacteria and release pro-inflammatory products. Second, phages can promote the expansion of CD8+ and IFN-γ-producing CD4+ T cell populations in lymph nodes in a TLR9/MyD88-dependent manner [24] (Figure 4C). After infection by Staphylococcus aureus or Pseudomonas aeruginosa phages, peripheral blood monocytes exhibited a transcriptional response and notably enhanced transcription of IL-1, IL-6, and TNF [81]. The filamentous phages in the chronic human wounds infected by P. aeruginosa triggered TLR3 activation and production of type I IFN in immune cells. In turn, this type of cytokines (IFNα and IFNβ) inhibited TNF production by macrophages, thereby impairing phagocytosis and bacterial clearance and delaying wound healing [82]. Despite these advances in understanding phages’ interaction with bacteria, it is still very early to understand the arms race. In the future, more work may focus on the influence of enteric phages on the composition of the bacterial microbiota that affects phage fitness and pathogenesis.

Collectively, the interactions among viruses, bacteria, and the host immune system are being studied more commonly. However, in many cases, causality and molecular mechanisms remain incompletely understood and should be explored more broadly mechanistically.

5. Phage-based therapeutics in intestinal inflammation and cancer

Phage-based therapeutics (also known as phage therapy) have existed for around 100 years [83]. The therapeutics for infectious diseases were almost abandoned in most western countries 60 years ago due to the unpredictable outcomes and dramatically improved therapy with newly discovered antibiotics [83]. However, in recent decades the abuse of antibiotics has caused the surge of antibiotic resistance. Due to the lack of effective treatments and the rapid evolution of bacterial resistance to antibiotics, phage therapies have found renewed enthusiasm as alternative approaches for multi-drug resistance bacterial infection, mostly in very severe cases. Therefore, there are a growing number of clinical reports and studies on the use of phage therapy to treat fatal bacterial infection [84] or other comorbidity diseases, such as cystic fibrosis and chronic obstructive pulmonary disease [85, 86].

Not surprisingly, because certain pathogenic bacteria are associated with both IBD and CRC, several clinical trials with phage-related therapy are ongoing in treating colitis and CRC [26, 87]. Phage therapy may be advantageous in microbiome manipulation with the high specificity of phages targeting single bacterial species [88]. For example, phages are being tested against Clostridioides difficile in UC, adherent invasive Escherichia coli in CD, and Fusobacterium nucleatum in colorectal cancer [21].

5.1. Phage-based therapy in IBD

Adherent-invasive E. coli (AIEC) might have a causal role in the pathogenesis of CD [89]. Galtier et al. isolated three phages targeting AIEC from wastewater, which can reduce the ileal and colonic colonization of AIEC and mitigate the symptoms of DSS-induced colitis in mice. This work provided a new therapeutic option in patients with CD [90]. A phase 2, double-blind, randomized, placebo-controlled clinical trial (NCT03808103) is enrolling 30 CD patients to evaluate the effectiveness of the AIEC-specific bacteriophage cocktail (EcoActive) on disease activity, inflammatory markers, and AIEC load.

5.2. Phage-based therapy in CRC

Fusobacterium nucleatum is implicated in the pathogenesis of CRC [91, 92]. Phages targeting F. nucleatum have also entered clinical trials to utilize these viruses to treat CRC and reduce the cancer burdens [93]. However, this single targeting also presents a challenge in using phages for emerging bacterial strains that have no available specific and effective phages, which can be developed by isolating phages from the wild [94] or by engineering to generate multi-targeting universal phages [95]. Gogokhia et al. [24] reported that Caudovirales phages isolated from active UC patients inhibited the growth of cancer-causing adherent invasive E. coli and suppressed intestinal tumor growth in a mouse model. Additionally, phages encode a depolymerase that enables them to degrade biofilms and access the residing organisms [96, 97]. Phages could also be engineered to carry additional therapeutic advantages [98]. For instance, using azide-modified phages linked to irinotecan-dextran nanoparticles for treatment of mice bearing CT26 colorectal carcinomas decreased Fusobacterium spp. levels and effectively suppressed tumor growth [21].

These above studies highlight the need to indicate the potential therapeutic role of phage therapy in colitis and CRC. The major limitation that phages usually have a narrow target range for bacteria may be solved by engineering multi-valent, broad targeting artificial phages, or stock more wild-derived phages for clinical applications. Moreover, randomized controlled clinical trials on an increased scale are required to expand phage therapy applications in humans for treating IBD and CRC.

6. Oncolytic viruses for CRC treatment

Cancer virotherapy is immunotherapy based on oncolytic viruses (OVs) that modulate the tumor microenvironments (TME) to reverse the immunosuppressive states and subsequently stimulating anti-tumor immunity [99, 100]. OVs are natural or genetically modified viruses designed to target and kill cancer cells without apparent damage to normal cells [101]. Thus far, various DNA and RNA OVs are rapidly emerging as novel therapeutic approaches against cancer, including herpes simplex virus (HSV), vaccinia virus (VAC), adenovirus (AdV), reovirus (RV), measles virus (MeV), etc. [102]. OVs are currently optimized by genetic modification or combination with other strategies to provide greater specificity and efficacy against tumors without harming healthy cells [103]. The ongoing clinical trials in patients with CRC are summarized in Table 2.

Table 2.

Clinical trials using oncolytic viruses for treating CRC patients.

Viral families	Viral species	Virus	Transgene	Condition	Administration	Combination	Status	Phase	Identifier1	Ref
dsDNA	VAC	Pexa-Vec (JX-594)	GM-CSF	CRC	i.v.	–	Completed	I	NCT01469611
dsDNA	VAC	Pexa-Vec (JX-594)	GM-CSF	CRC	i.v. or i.t.	–	Completed	II	NCT01329809
dsDNA	VAC	Pexa-Vec (JX-594)	GM-CSF	CRC	i.v.	Irinotecan	Completed	I/II	NCT01394939
dsDNA	VAC	Pexa-Vec (JX-594)	GM-CSF	CRC	i.v.	–	Completed	I	NCT01380600
dsDNA	VAC	Pexa-Vec (JX-594)	GM-CSF	CRC	i.v.	anti-PD-L1, anti-CTLA-4	Active	I/II	NCT03206073	[105]
dsDNA	VAC	TBio-6517	FLT3 ligand, IL-12, anti-CTLA-4	Solid tumors, TNBC, MSS-CRC	i.t.	anti-PD-1	Recruiting	I/II	NCT04301011
dsDNA	AdV	LOAd703	CD40L, 4-1 BBL	Pancreatic, ovarian, biliary cancer, CRC	i.t.	Chemotherapy	Recruiting	I/II	NCT03225989	[108]
dsDNA	AdV	EnAd (ColoAd1)	–	Solid tumors of epithelial origin, metastatic CRC, and bladder cancer	i.v.	–	Completed	I/II	NCT02028442	[111]
dsDNA	AdV	EnAd (ColoAd1)	–	Resectable CRC, non-small cell lung, bladder, renal cell cancer	i.v. or i.t.	–	Completed	I	NCT02053220	[112]
dsDNA	AdV	EnAd (ColoAd1)	–	CRC, squamous cell carcinoma of the head and neck	i.v.	anti-PD-1	Active	I	NCT02636036
dsDNA	AdV	EnAd (ColoAd1)	–	Locally advanced rectal cancer		Capecitabine, radiotherapy	Recruiting	I	NCT03916510	[110]
dsDNA	HSV	T-VEC	GM-CSF	Metastatic TNBC & CRC	i.a.	anti-PD-L1	Active	I	NCT03256344	[106, 107]
dsDNA	HSV	NV1020	–	CRC, liver neoplasms	i.a.	–	Completed	I/II	NCT00149396	[121]
dsDNA	HSV	NV1020	–	CRC, metastatic cancer	i.a.	–	Completed	I	NCT00012155	[122]
dsDNA	HSV	OH2	GM-CSF	Solid tumor, gastrointestinal cancer	i.v.	anti-PD-1, Irinotecan	Recruiting	I/II	NCT03866525
dsDNA	HSV	ONCR-177	IL-12, CCL4, FLT3L, anti-PD-L1, anti-CTLA-4	CRC2	i.t.	anti-PD-1	Recruiting	I	NCT04348916	[109]
Segmented dsRNA	RV	Reolysin	–	KRAS mutant metastatic CRC	i.v.	Irinotecan, Leucovorin, 5-FU, anti-VEGF	Completed	I	NCT01274624	[113]
ssRNA	MeV	TMV-018	Cytosine deaminase	Gastrointestinal cancer	i.t.	5-FC, anti-PD-1	Withdrawn	I	NCT04195373

5-FC, 5-fluorocytosine; 5-FU, fluorouracil; AdV, adenovirus; CTLA-4, cytotoxic T lymphocyte antigen-4; dsDNA, double-stranded DNA; dsRNA, double-stranded RNA; GM-CSF, granulocyte-macrophage colony-stimulating factor; HSV, herpes simplex virus; i.a., intrahepatic arterial; i.t., intratumoral; i.v., intravenous; MeV, measles virus; MSS-CRC, microsatellite stable colorectal cancer; PD-1, programmed cell death protein 1; PD-L1, programmed death ligand-1; Ref, reference; RV, reovirus; ssRNA, single-stranded RNA; TNBC, triple-negative breast cancer; VAC, vaccinia virus; VEGF, vascular endothelial growth factor; T-VEC, talimogene laherparepvec.

¹Clinicaltrials.gov Identifier.

²Melanoma, solid tumor, squamous cell carcinoma of head and neck, breast cancer, advanced solid tumor, TNBC, CRC, non-melanoma skin cancer, liver metastases. Source: clinicaltrials.gov; assessed May 2021.

OVs can act as cancer vaccines to increase tumor-specific T cell response. OVs can also be armed with immunostimulatory molecules (i.e., granulocyte-macrophage colony-stimulating factor, GM-CSF) to improve their immune-activating characteristic. The OVs armed with GM-CSF facilitate dendritic cell (DC) migration and maturation, eventually leading to enhanced priming of T cell responses [104], such as talimogene laherparepvec (T-VEC) and pexastimogene devacirepvec (Pexa-Vec or JX-594) [105]. Talimogene laherparepvec (T-VEC) system (HSV) is a well-known, therapeutically modified virus. T-VEC has been approved by both the US Food and Drug Administration (FDA) and the European Medicines Agency to treat metastatic melanomas [106, 107]. T-VEC is currently under clinical trial for CRC [106, 107]. Pexa-Vec is a VAC modified to encode GM-CSF and β-galactosidase to inactivate the viral thymidine kinase gene and is currently undergoing multiple clinical trials [105].

IL-12, a major orchestrator of Th1-type immune response against cancer, is another cytokine used for arming OVs, such as TBio-6517 (Phase I/II) and ONCR-177 (Phase I) [108, 109]. LOAd703 (Phase I/II) is a double-armed ADV combined of two tumor necrosis factor receptor (TNF) family ligands CD40L and 4-1BBL, which stimulates T cell expansion, acquisition of effector function, survival, and development of T cell memory [108].

Enadenotucirev (EnAd; ColoAd1) is a complex chimeric virus resulting from recombination between different adenovirus serotypes, and is currently under clinical investigations [110–112]. Unmodified viruses, such as the reovirus Pelareorep (Reolysin), are also being examined in clinical trials for CRC treatment [113].

In addition, OVs can be engineered in combination with immune checkpoint inhibitors (ICIs) or cytotoxic agents to achieve the most potent cancer immunotherapy through a synergistic mechanism, such as Pexa-Vec with Tremelimumab (anti-CTLA-4) and Durvalumab (against PD-L1), TBio-6517 with Pembrolizumab (anti-PD-1), ColoAd1 with nivolumab (anti-PD-1), T-VEC with Atezolizumab (anti-PD-L1), OH2 with HX008 (anti-PD-1), and ONCR-177 with Pembrolizumab. Through these multi-pronged research and clinical trials, we envision that some OVs may be able to earn FDA approval for clinical applications in cancer treatment successfully.

7. Emerging technologies for investigating enteric virome

The emergence of high throughput metagenomic sequencing technology allows us to understand the complexity and richness of human gut bacteriophage and various viruses’ populations [27, 114]. However, compared to the bacterial component of the intestinal microbiome, the enteric virome has been almost ignored, largely due to the limited tools available for virus identification and classification [115]. It might also be due to the significant interest in the bacterial microbiome per se, so the importance of virome is somewhat ignored. In addition, viruses exist in several genetic forms that differ by the backbone nucleic acid (RNA or DNA), the feature of strands (positive or negative sense), and the number of strands (single vs. double strands). The complexity of viruses poses challenges for library preparation and sequencing strategies. At present, it is estimated that merely 1% of the virome has been sequenced, and the number of unclassifiable sequences (taxonomically or functionally) ranges from 60% to 90% (called ‘viral dark matter’), which is awaiting characterized [116].

VLPs fractions can be examined using transmission electron microscopy (TEM) [117], metagenomic sequencing [31], or high-throughput short-read-based technologies (Roche 454, Illumina platforms, and Ion Torrent platforms) [29]. Recently, two long-read sequencing technologies (Pacific Biosciences and Oxford Nanopore) have been developed. These technologies can assist in the construction of large novel viral genomes, obtain information on methylation patterns [118], and study population structure at a single virion level [119]. The virus metagenomic analysis workflow includes quality control, filtering and trimming of non-target reads, assembly of reads into contigs, removal of bacterial contamination, alignments to viral genomes in viral databases, and downstream analysis. A number of software and databases have been specifically designed to process high-throughput virome sequencing data (Table 3). These tools should be used after careful consideration of sample type and scientific questions. Therefore, a critical limitation is whether to have an expert bioinformatician and the appropriate hardware necessary to perform such extensive and time-consuming analysis.

Table 3.

Selected methods and databases for viral metagenomic analysis.

Tool	Description	Reference	URL
Quality control and filtering non-target reads
Trimmomatic	A variety of useful trimming tasks for Illumina paired-end and single-ended data.	[123]	https://github.com/timflutre/trimmomatic
fastp	Providing fast all-in-one preprocessing for fastq files, including quality control, trimming adapters, filtering by quality, and read pruning.	[124]	https://github.com/OpenGene/fastp
cutadapt	Finding and removing adapter sequences, primers, poly-A tails, and other types of unwanted sequences from high-throughput sequencing reads.	[125]	https://github.com/marcelm/cutadapt/
FastQC	An application that takes a fastq file and runs a series of tests on it to generate a comprehensive QC report.	[126]	http://www.bioinformatics.babraham.ac.uk/projects/download.html#fastqc
Short Read Assembly
de novo assembly and mapping
SPAdes	An assembly toolkit containing various assembly pipelines for high-throughput sequencing data.	[127]	https://github.com/ablab/spades
VirFinder	A novel k-mer based tool for identifying viral sequences from assembled metagenomic data.	[128]	https://github.com/jessieren/VirFinder
SunBeam	An extensible pipeline for analyzing metagenomic sequencing experiments.	[129]	https://github.com/sunbeam-labs/sunbeam
Reference-based mapping
Bowtie 2	An ultrafast and memory-efficient tool for aligning sequencing reads to long reference sequences.	[130]	https://github.com/BenLangmead/bowtie2
bwa	A software package for mapping DNA sequences against a large reference genome.	[131]	https://github.com/lh3/bwa
Kraken2	A taxonomic classification system using exact k-mer matches to achieve high accuracy and fast classification speeds.	[132]	https://github.com/DerrickWood/kraken2
Pavian	A comprehensive visualization program that can compare Kraken 2 classifications across multiple samples.	[133]	https://github.com/fbreitwieser/pavian
Bracken	Allows users to estimate relative abundances within a specific sample from Kraken 2 classification results.	[134]	https://github.com/jenniferlu717/Bracken
Downstream analysis
mauve	A system for constructing multiple genome alignments in the presence of large-scale evolutionary events such as rearrangement and inversion.	[135]	http://darlinglab.org/mauve/mauve.html
PyANI	Pairwise average nucleotide identity between genomes.		https://github.com/widdowquinn/pyani
Gview	A Java package used to display and navigate bacterial genomes.	[136]	https://github.com/phac-nml/gview-wiki/wiki
vegan	Ordination methods, diversity analysis, and other functions for community and vegetation ecologists.		https://github.com/vegandevs/vegan
ape	Phylogenetic and evolutionary analysis of DNA and protein sequence data.	[137]	http://ape-package.ird.fr/
Databases
NCBI RefSeq Viral Genomes	Provide viral genome sequence data and related information.		https://www.ncbi.nlm.nih.gov/genome/viruses/
IMG/VR	Provide information of genomes of cultivated and uncultivated viruses.	[138]	https://img.jgi.doe.gov/cgi-bin/vr/main.cgi
pVOGs	Provides access to the most recent database 9,518 orthologous groups shared among nearly 3,000 thousand complete genomes of viruses that infect bacteria and archaea (Prokaryotic Virus Orthologous Groups, or pVOGs).	[139]	http://dmk-brain.ecn.uiowa.edu/pVOGs/

Apart from the virome metagenomic sequencing, additional future complementary approaches may include gut viral metatranscriptomics (RNA-seq) and viral metaproteomics [29, 120]. The technologies mentioned above and other future development may well shape virome research to probe the pathogenesis of gut virome and its potential beneficial influence in humans, providing insight into the design of effective virome-based therapy for intestine and colorectal diseases.

8. Challenges and future directions

Despite these relatively rapid advances, there are still major gaps in our understanding of the nature of enteric viruses.

1. Although the involvement of viruses in the development of colitis and CRC has become increasingly evident, the current concepts and observations need further testing and verification.

2. With the advent of many new technologies, such as nucleic acid sequencing, omics analysis, and bioinformatics pipelines, the characterization of the gut virome and its role in the development of pathological conditions is still at an early stage.

3. The virus database is ralatively small and incomplete (compared to the bacterial genome database), limiting our ability to dissect the detail of mucosal viromes in health and disease.

4. Virology research regarding human health faces considerable inter-individual variability that may be affected by many factors, such as age, sex, ethnicity, geography, diet, and sample collection, storage, and processing.

In summary, there are many important advances in virome research, basic understanding, disease relevance, and clinical usage/therapy; there are also multiple hurdles. These obstacles mentioned above must be addressed appropriately before phage therapy and OVs are approved for broad-scope clinical application. Despite these concerns, viral therapeutics may be worth exploring and may have huge potential through molecular engineering.

Besides providing a basic concept and summarizing the connection of virome to IBD and CRC, we also discussed that phages are a promising therapeutic tool against pathogenic bacteria for human inflammatory bowel disease and colorectal cancer. Randomized, placebo-controlled trials with phage therapy are already ongoing in IBD and CRC. In addition, there is an urgent need to design new approaches and invent methods in virus isolation, metagenomics, enrichment culture, and bioinformatics tools to improve our ability to define and characterize viruses in the future. Further large-scale longitudinal and long-term follow-up prospective studies, as well as in-depth experimental investigation and validation are necessary. The key direction in this field is to determine the dynamic relationship (e.g., causal role) between the intestinal microbiota and innate and adaptive immunity. Furthermore, phage therapy may be applied together with bacterial or other microbiota components (via FMT, pre- and probiotics) to alter virulence and immunogenicity, effectively taming the exacerbated inflammatory disease in humans and improving human health.

Abbreviations

5-FC

5-fluorocytosine

5-FU

fluorouracil

AdV

adenovirus

ALEC

Adherent-invasive Escherichia coli

CAC

colitis-associated colorectal cancer

CD

Crohn’s disease

cDNA

complementary DNA

CMV

cytomegalovirus

CRC

colorectal cancer

CTLA-4

cytotoxic T lymphocyte antigen-4

dsDNA

double-stranded DNA

dsRNA

double-stranded RNA

EnAd

Enadenotucirev

FDA

Food and Drug Administration

gDNA

genomic DNA

GM-CSF

granulocyte-macrophage colony-stimulating factor

HBV

hepatitis B virus

HCV

hepatitis C virus

HPV

human papillomavirus

HSV

herpes simplex type virus

i.a.

intrahepatic arterial

i.t.

intratumoral

i.v.

intravenous

IARC

International Agency for Research on Cancer

IBD

inflammatory bowel disease

ICI

immune checkpoint inhibitor

IEC

intestinal epithelial cell

JCV

John Cunningham virus

LPS

lipopolysaccharide

MeV

measles virus

MMTV

Mouse mammary tumor virus

MNoV

Murine norovirus

MSS-CRC

microsatellite stable colorectal cancer

OV

oncolytic virus

PD-1

programmed cell death protein 1

PD-L1

programmed death ligand-1

Pexa-Vec

pexastimogene devacirepvec

PV

poliovirus

Ref

reference

RV

reovirus

ssDNA

single-stranded DNA

ssRNA

single-stranded RNA

T-VEC

Talimogene laherparepvec

TEM

transmission electron microscopy

TME

tumor microenvironment

TNBC

triple-negative breast cancer

UC

ulcerative colitis

VAC

vaccinia virus

VEGF

vascular endothelial growth factor

VLP

virus-like particle