The microbiota formed by microorganisms and residing in the gastrointestinal tract is referred to as “intestinal microbiota” or “gut microbiota” [1
]. The microbiota affects various aspects of human health, including providing nutrients and vitamins, protecting against pathogens, epithelial mucosa homeostasis, and immune system development [3
]. Microbial dysbiosis has been linked to various metabolic and inflammatory diseases, such as diabetes mellitus, hypertension, inflammatory bowel disease, and obesity [1
]. Growing evidence not only implies that chemotherapeutics affect the intestinal microbial composition but also that multidirectional interactions between the gut microbiota and the host immune system may influence development and progression of chemotherapy-induced intestinal inflammation [4
Gastrointestinal toxicity is a severe, dose-limiting, toxic side effect of chemotherapeutics in cancer patients. Mucositis and diarrhea are a primary outcome of intestinal toxicity and result in an increased hospitalization duration and infection risk as well as reduced anti-neoplastic treatment efficacy, leading to reduced survival and a substantial burden on Medicare [7
]. The development of chemotherapy-induced mucositis involves a complex and dynamic array of biological events occurring in five interconnected phases, i.e., initiation, upregulation and generation of messenger signals, signal amplification, ulceration, and healing [7
]. Several pathogenic elements, including direct toxicity, change in the bowel microbial flora, oxidative stress, apoptosis, hypo-proliferation, and abnormal inflammation, are involved. Unfortunately, no well-established or effective therapeutic strategies are currently available for the management of chemotherapy-induced intestinal mucositis [8
Fecal microbiota transplant (FMT) is the transfer of fecal material, including bacteria and natural anti-bacterials, from a healthy individual into a diseased recipient. The manipulation of gut microbiota by FMT has been performed in various disease models and clinical trials and has received much public attention [10
]. Due to the high cure rate (over 90%) and the rarity of side effects associated with it, FMT has been considered a potentially life-saving, “last chance” option for the treatment of recurrent Clostridium difficile
]. This has generated interest in its use for the treatment of other gastrointestinal and even inflammatory diseases. Several studies have reported that the infusion of a fecal suspension from healthy individuals to patients with inflammatory bowel disease or irritable bowel syndrome resulted in clinical improvement and remission [12
]. It has been also suggested that FMT shows promising therapeutic potential for the treatment of diabetes, obesity, non-alcoholic fatty liver disease, and even cardiovascular disease [12
]. Although the underlying molecular and cellular mechanism for FMT remains unclear, it may involve the direct interaction of donor gut microbiota with that of the host, which subsequently mediates the effects observed on host physiology, gut mucosal barrier, and immune system [14
]. Thus, FMT may be effective at manipulating gut microbiota and ameliorating inflammation and severity of mucositis induced by chemotherapy.
Several concerns have been raised regarding the safety of FMT therapy. Specifically, adverse events occur often and should be carefully monitored throughout the FMT process [15
]. In theory, it may be possible to transmit the potentially harmful traits of microbiota, thereby enabling the transmission of occult infections [16
]. Further, transient adverse events, including mild fever, abdominal pain, diarrhea, exhaustion, flatulence, and fatigue, were reported following FMT [16
]. These adverse effects were self-limiting. FMT was proven safe in the short-term follow-up period; however, long-term safety must be assessed [16
]. Furthermore, though evidence regarding the efficacy and the safety of FMT in immunocompromised patients has begun to emerge, data pertaining to cancer patients treated with anti-neoplastic agents remain scarcely available [18
Regimens based on 5-fluorouracil (5-FU) with the cytotoxic agent oxaliplatin, FOLFOX (5-fluorouracil, leucovorin, and oxaliplatin) have been widely used in standard chemotherapy for advanced and metastatic colorectal cancer treatment [21
]. The combination therapy of oxaliplatin with 5-FU potentiates gastrointestinal toxicity in clinical studies; however, the underlying mechanism remains unclear [7
]. In our previous study, we reported on a colorectal cancer murine model with FOLFOX-induced intestinal mucositis [24
]. It involves changes in gut microbiota and may be “driven” by NF-κB expression, which subsequently induces the generation of apoptotic signals and pro-inflammatory cytokines, which contribute to gastrointestinal injury [24
]. By modulating the gut microbiota and the pro-inflammatory responses, probiotics mitigated FOLFOX-induced mucositis [24
Accordingly, in the current study, we sought to investigate the effect and the safety of FMT on 5-FU-based chemotherapy (FOLFOX)-induced intestinal mucosal injury in mice with colon cancer. We performed FMT using stool from healthy wild-type donor mice to restore microbial diversity and composition. The possible mechanisms of action for FMT were also elucidated.
This is the first study to our knowledge to report on the potential and the safety of use of FMT for suppressing FOLFOX-induced mucositis in a colorectal cancer mouse model in vivo. FOLFOX administration significantly prevented tumor growth in colon cancer-bearing mice. FMT ameliorated FOLFOX-induced severe diarrhea, bacterial translocation, and intestinal mucosal injury and improved long-term survival associated with FOLFOX administration in CT26 colorectal cancer-bearing mice. Furthermore, FMT reduced FOLFOX-induced intestinal mucosal inflammation and barrier integrity disruption. This was characterized by immunohistological changes and expression of tight junction proteins. FOLFOX-induced intestinal mucosal apoptosis, changes in fecal gut microbiota, and TLR expression were also ameliorated by FMT. Therefore, these results indicate that FMT may have clinical potential for the management of chemotherapy-induced intestinal dysbiosis and toxicity.
The transplantation of fecal microbiota was performed in order to manipulate the gut microbiota in various inflammatory disease models in vivo [2
]; however, few studies have investigated the effect of FMT on gastrointestinal mucositis resulting from combined 5-FU and oxaliplatin chemotherapy in animal models [2
]. The evidence supporting the use of FMT for achieving improved chemotherapy toxicities is inconclusive, and studies investigating this topic are ongoing [2
]. Li et al. revealed that fecal transplantation from healthy mice might alleviate weight loss, colon shortening, and intestinal mucositis induced by 5-FU in healthy BALB/c mice [30
]. Our previous study reported that oral probiotic Lcr35
prevented FOLFOX-induced intestinal mucositis in colorectal cancer-bearing mice [24
]. The colorectal cancer-bearing mice model mimicked side effects of chemotherapy-associated gastrointestinal toxicity and diarrhea. Here, we showed that oral FMT significantly attenuated diarrhea and improved diarrhea scores in the FOLFOX groups. Furthermore, the histological analysis of the intestinal mucosal injury caused by FOLFOX in the mouse model was prevented by FMT. Our study revealed that FMT may ameliorate the severity of mucositis induced by chemotherapy.
In mucosal barrier integrity, primarily the mucus layer, the tight junctions, and the intestinal epithelial cells, dysfunction plays an important role in the pathogenesis of chemotherapy-induced gut toxicity and associated diarrhea [31
]. This mucus layer is composed of mucins, which are diverse, complex glycoproteins secreted by the goblet cells. Tight junctions seal the paracellular space of the intestinal epithelium and maintain structure and function of the intestinal mucosal barrier, which is disrupted under inflammation conditions [32
]. Tight junctions are composed of core proteins such as the transmembrane proteins, including claudins (CLDNs), occludin, and junctional adhesion molecules (JAMs), and these molecules bind directly to periplasmic (intracellular) scaffolding proteins such as zonula occludens (ZOs). ZOs are the most important components for the construction of a constitutive barrier of epithelial cells, and they regulate the permeability of the barrier by tightly sealing the cell–cell junctions [34
]. Several potential markers of the intestinal barrier, the ileum, or the colon can be assessed in mice, including mucus thickness, mRNA expression, and levels of Muc-2 and junction molecules (JAM-A, ZO-1, occludin, claudin-1, -2, and -5) [35
]. Chemotherapeutic drugs could increase intestinal epithelial barrier permeability by reducing tight junction expression in vivo. Song et al. investigated the 5-fluorouracil-induced changes in intestinal integrity biomarkers in BALB/C mice. They showed that the expression levels of occludin and claudin-1 were significantly reduced, while that of ZO-1 was unchanged in the small intestines of mice after injecting 5-FU once as compared to those of the control group [36
]. Li et al. revealed that, while levels of the tight junction protein occludin were reduced, levels of ZO-1 and junctional adhesion molecule-A (JAM-A) were increased in colonic tissues of mice by injecting 5-FU once daily for three days [30
]. The effects of chemotherapeutic drugs such as 5-FU on the intestinal epithelial barrier via the expression of tight junction proteins are reported inconsistently [30
] This might be attributable to differences in regimens, experimental protocols, or animal models used. Our study also revealed that FMT prevented FOLFOX-induced villus damage by significantly affecting the differentiation of mucin-filled goblet cells in the intestine. Thus, FMT prevented FOLFOX-induced dysfunction in mucosal barrier integrity and affected goblet cell differentiation and intercellular scaffolding tight junction expression of proteins such as ZO-1. This suggests that FMT could reduce FOLFOX-induced mucosal barrier damage in colorectal tumor-bearing mice.
Current studies on the safety of FMT in patients or animal models with tumors and in the context of treatment with anti-neoplastic agents are limited [16
]. Cancer patients not only have a high mortality because of individual episodes of sepsis but suffer disproportionately from sepsis as compared to the general population [38
]. The increased risk of sepsis in cancer patients has been attributed to myelosuppression, mucositis, and the use of central venous catheters [39
]. Gut pathogens are an important cause of sepsis in cancer patients. Damage to the gut epithelium after chemotherapy, cancer itself, and bacterial overgrowth contribute to bacterial translocation, making those receiving chemotherapy particularly vulnerable to bloodstream infections caused by enteric bacteria [40
]. Because feces act as mediators between the donor and the recipient, FMT has the potential to transmit occult infections, even when donor screening is performed stringently [16
]. Analyses made during a short-term follow-up period showed that it was safe to perform FMT; however, recent reports show that adverse events occurring after FMT were reportedly associated with the mortality rate for Clostridium difficile
infections, including septic shock with decompensated toxic megacolon and fatal aspiration pneumonia [16
]. Therefore, the safety of the FMT process continues to limit its use in immunocompromised and cancer patients treated with anti-neoplastic agents, and its long-term side effects, if any, remain unknown [16
]. However, FMT might reportedly be able to correct dysbiosis and prevent bacterial translocation and sepsis and improve survival. For instance, Li et at showed that, in a mouse model, FMT could not only reverse bacterial translocation but could also improve the survival period associated with experimental necrotizing enterocolitis [28
]. FMT caused the reversal of survival in a BALB/c mouse model of ulcerative colitis [27
]. FMT attenuated radiation-induced toxicity, which eventually led to death in a mouse model [29
]. Our studies showed that, in a colorectal cancer-bearing mouse model, at necropsy, none of the bacteria were detected to the bloodstream in all experimental groups; however, bacterial translocation to mesenteric lymph nodes, liver, and spleen was more frequent in the FOLFOX-treated group than in the saline group. FMT tended to reduce bacterial translocation in the FOLFOX-treated group. In addition, there is a potential risk of treatment-related mortality in cancer patients receiving FOLFOX [41
]. In our chemotherapy-related mouse model with gastrointestinal toxicity, we also observed that FOLFOX administration reduced mouse survival. FMT increased the survival rate of FOLFOX-treated mice in the long-term experimental study, suggesting that FMT might safely provide protection against chemotherapy-induced toxicity and mortality in this colorectal cancer-bearing mouse model treated with anti-neoplastic agents.
TLRs act as sensors for microbial infection and are critical for the initiation of inflammation and immune defense responses [42
]. TLRs recognize ligand from fungi, viruses, and bacteria. Bacterial ligands are not only unique to pathogens but are found in all bacteria and are produced by symbiotic microorganisms [43
]. During chemotherapy, chemotherapeutic agents can reduce the number and the diversity of microbiota and cause dysbiosis [6
]. TLRs can sense dysbiosis and danger signals from dead or injured host cells. Upon activation, each TLR differentially recruits members of a set of TIR domain-containing adaptors, such as MyD88, TRIF, TIRAP/MAL, or TRAM. Collectively, depending on adaptor usage, TLR signaling is largely divided into two pathways, i.e., the MyD88-dependent and the TRIF-dependent pathways. The subsequent transcriptional activation of TLR target genes encoding pro-inflammatory and anti-inflammatory cytokines, chemokines, effector molecules, and type I interferons initiates the activation of antigen-specific and non-specific adaptive immune responses [44
]. MyD88 is utilized by all TLRs and activates NF-κB and MAPKs for the subsequent induction of inflammatory cytokine genes, such as granulocyte colony-stimulating factor, interleukin-1β, IL-6, and TNF-α. Thus, MyD88 and downstream NF-κB are the key signaling molecules involved in the MyD88-dependent TLR signaling pathway [45
]. Previous studies have shown that intestinal epithelial cells expressed several TLRs, including TLR1, TLR2, TLR4, TLR5, and TLR6 in vitro and in vivo, and that these TLRs were expressed on the cell surface and mainly recognized microbial membrane components, such as lipids, lipoproteins, and proteins [25
]. TLR4 recognizes bacterial lipopolysaccharide (LPS). TLR5 recognizes bacterial flagellin [50
]. The heterodimers formed by combinations of TLR1, TLR2, and TLR6 recognize a wide variety of pathogen-associated molecular patterns, including lipoproteins, peptidoglycans, lipoteichoic acids, zymosan, mannan, and tGPI-mucin [25
]. In our study, FOLFOX treatment might have disturbed the original balance of gut microbiota and significantly upregulated the mRNA expression of TLRs (TLR1, TLR2, TLR4, TLR5, and TLR6) and key signal molecule MyD88 in the intestine of mice. FMT significantly suppressed the expression of TLRs and MyD88 in the FOLFOX group. Thus, our data suggest that FMT can suppress the effects of FOLFOX and induce the MyD88-dependent TLR signaling pathway; this indicates that intestinal microbiota of recipient mice in the FOLFOX group and homeostasis were reestablished.
The mechanisms by which the clinical benefits of FMT are achieved, including in inflammatory disease patients, are not completely understood. Several potential mechanisms might primarily include restoration of the intestinal microenvironment, direct interaction of donor gut microbiota with the host physiology, gut mucosal barrier, and immune system [13
]. NF-κB is a transcription factor that regulates the expression of numerous genes that are critical for survival and control inflammation, cell growth, apoptosis, and the cell cycle [9
]. NF-κB plays a central role in the pathobiology of the five-phase model for the development of chemotherapy-induced intestinal mucositis and acts as a “gatekeeper” for various pathways [9
]. Its activation is induced by anti-neoplastic agents, such as 5-FU, which are thought to elicit inflammatory and apoptotic responses in the intestine [9
]. NF-κB activation triggers the activation of multiple signaling pathways for the synthesis of different pro-inflammatory cytokines involved in various stages of mucositis, including TNF, IL-1β, and IL-6 [9
]. NF-κB is also the key signaling molecule of the TLR signaling pathway involved in the induction of inflammatory cytokine genes (such as IL-1β, IL-6, and TNF-α) [45
]. In addition, a number of studies have demonstrated a change in tight junction permeability attributable to TNF-α, IL-1β, and IL-6, and these changes in tight junctions were involved in the functioning of the NF-κB signaling pathway [52
]. In this study, immunohistological analysis indicated that FMT decreased FOLFOX-induced NF-κB activity in the intestine and reduced FOLFOX-induced apoptosis. We also showed that serum levels of IL-1β and IL-6 were significant, and that the TNF-α level tended to be increased in the FOLFOX group, while FMT abrogated the effect of IL-6. These findings suggest that the inhibition of NF-κB activity by FMT might result in the suppression of inflammation and the sequential amelioration of FOLFOX-induced mucosal barrier damage and apoptosis in the intestine. Accordingly, our data suggest that, by modulating the composition of the gut flora, FMT altered FOLFOX-induced changes in the gut microbiota and influenced the pathogenesis of mucositis via the gut microbiota-TLR-NF-κB signaling pathway in colorectal carcinoma-implanted mice.
Gut microbiota homeostasis has been manipulated via FMT for treating various local and systemic inflammatory diseases linked to gut microbiota models and clinical trials, such as inflammatory bowel disease, radiotherapy-induced diarrhea, obesity, and diabetes [2
]. Chemotherapy is associated with changes in microbial diversity [4
]. This change in microbial diversity coincides with the development of severe chemotherapy-induced mucositis; commensal intestinal bacteria could be significantly involved in the pathogenesis of mucositis [54
]. Some studies demonstrated that the gut microbiota was actively involved in the pathogenesis of 5-FU induced intestinal mucositis, which suggests that 5-FU induced intestinal mucositis could be potentially attenuated by disturbing the homeostasis of gut microbiota [2
]. Accordingly, the disturbance in the homeostasis of gut microbiota could have therapeutic effects against chemotherapy-induced mucositis. Here, FOLFOX changed the composition of gut microbiota, and the oral FMT altered this compositional change. Further taxonomic analyses at the phylum level indicated that FOLFOX significantly increased the abundance ratio of Firmicutes/Bacteroidetes (F/B). These changes were restored by FMT. Bacteroidetes and Firmicutes are the predominant phyla in humans and mice [6
]. The balance between these two phyla (F/B ratio) appears to be critical for the regulation of radiotherapy or chemotherapy-related mucositis [6
]. Le Bastard et al. showed that 5-FU-induced gut dysbiosis, richness, and diversity were restored by FMT in a healthy mouse model [59
]. In tumor-bearing mice, our study revealed that the alpha diversity indexes, including the Shannon (community diversity) index, were determined, and no significant differences were found with regard to the microbiota of saline, FMT, FOLFOX, and FOLFOX+FMT groups. However, beta diversity analysis using the Jaccard and the Bray–Curtis distance matrix analysis method showed FMT and FOLFOX treatment alone affected the composition of the fecal gut microbiota. Because a Jaccard analysis accounts merely for the presence/absence of OTUs, and a Bray–Curtis analysis is more sensitive to changes in OTU abundance [60
], the compositional changes in the fecal microbiota used for FMT in the FOLFOX+FMT group were probably influenced by the exclusion or the inclusion of certain species and changes in the abundance of present species.
Our study also has some limitations. We did not work in anaerobic conditions for preparation of frozen fecal material. The sample size of mice was small, and mice were co-housed, not single-housed, for this study. During co-housing, animals may feed on feces (also known as coprophagy) or ingest feces by self-grooming [61
]. It is a risk of transmission of gut microbiota between littermates [61
]. The study may need to be repeated and validated in different groups of mice single-housed in the future before clinical trials of FMT. We did not conduct an analysis of time-course and dose-dependent effects of FMT; it would have provided more information regarding their role in the pathogenesis of chemotherapy-induced mucositis. We did not conduct further in vivo studies of gene knockout or knockdown or in vitro studies, which would have elucidated the anti-mucositis mechanism. Our microbiota studies were small in size; only fecal samples, not colon content, were collected, and we did not identify the specific microorganisms associated with chemotherapy injuries and the functional roles of these microorganisms in the gastrointestinal tract. Clinical studies also need to be performed to demonstrate the beneficial effects of FMT and elucidate safety and effective regimens for the management of chemotherapy-induced mucositis.