Ulcerative colitis (UC) is a chronic idiopathic inflammatory bowel disorder, which is difficult to cure and easy to relapse. It has become a global disease with a high prevalence in western countries, and the growing incidence in newly industrialized countries (Ng et al., 2018). The exact pathogenesis of UC is thought to be a multi-factor disease, which is a result of the interaction between host susceptibility genes, environment, diet, immunity, intestinal barrier, and gut microbiota (Ramos and Papadakis, 2019). A large body of evidence has revealed the intestinal microbiome playing a critical role in the pathogenesis of UC, and has shown major shifts of intestinal flora in UC patients, such as reduced bacterial diversity, higher abundance of Proteobacteria and lower abundance of Firmicutes (Sokol and Seksik, 2010; Lane et al., 2017; Franzosa et al., 2019). However, the perturbation of specific microbes involved in UC pathogenesis have not been identified.

Several clinical trials have been performed to treat UC by manipulating the intestinal microbiota through probiotics, prebiotics, synbiotics, and antibiotics (Ghouri et al., 2014; Asto et al., 2019; Xi et al., 2021). However, evidence supporting significant therapeutic effects for UC remains limited. Fecal microbiota transplantation (FMT) is a powerful way to manipulate gut microbiota, which has been proven to be an effective treatment for Clostridium difficile infection (CDI). So far, only five randomized controlled trials (RCTs) studies were published regarding FMT treatment for UC (Supplementary Table 1), including four for adult patients and one for pediatric patients, and three of which reported higher remission rates [24%(Moayyedi et al., 2015), 27%(Paramsothy et al., 2017), and 32%(Costello et al., 2019)] in patients received high-intensity FMT compared to placebo (5, 8, and 9%, respectively). The pediatric RCT study also showed a higher composite clinical outcome of FMT group versus placebo group (91.7% vs. 50% at 6 weeks) (Pai and Popov, 2017; Pai et al., 2021). One RCT study reported a negative result in treating UC, with remission rates 30.4% vs. 20.0% compared to the control (p = 0.51) (Rossen et al., 2015). The uniform protocols of FMT administration have not been achieved, and there is a paucity of data on the long-term durability and safety of FMT in patients with active UC especially those with moderate to severe disease. In addition, most previous studies including RCTs performed FMT with high intensities, which increases the practical barriers. In this study, we assessed the long-term efficacy and safety of single-donor, low-frequency FMT in the treatment of UC, and analyzed the intestinal microbiota characteristics associated with different therapeutic outcomes.

Differences in FMT procedures might cause different therapeutic effects in various studies. Most previous studies including RCTs performed FMT with high intensities (pooled donors and frequent treatments over a short duration). FMT with a single donor and long interval between administrations could be considered as a low intensity regimen. Although UC patients benefit from FMT, concerns about the efficacy, safety, and durability of low-intensity FMT for UC treatment remains to be addressed. In this study, we performed FMT using stool suspensions from a single donor with time interval of 2 months for all UC patients, and achieved a remission rate of 22.58% (7/31) after the first FMT and 60% (12/20) after the second FMT. All the clinical indicators including Mayo scores, Endoscopic mayo scores, hematochezia scores, stool frequency, IBDQ were improved after the two FMT treatments. Thus, our results suggested that a low-intensity single donor FMT can lead to positive therapeutic results.

The long-term outcomes and health consequences of FMT remain to be established, with most studies reporting mixed results and without showing a sustained benefit of FMT (Brandt et al., 2012; Kump et al., 2013; Vermeire et al., 2016). Analysis with 109 UC patients received FMT from two clinical trials (NCT01790061, NCT02560727), 21.1% (23/109) and 25.7% (28/109) of patients maintained clinical response at 6 months after single FMT, and step multiple FMTs, respectively (Ding et al., 2019). In another study (Sood et al., 2019a), that the primary outcome of maintenance of clinical remission at 48 weeks was achieved in 27/31 [87.1%] patients who received FMT via colonoscopy every 8 weeks. In the present study, the longest remission time was 6 years which continued till this article was completed, and in 8 of the 12 patients (66.7%) in clinical remission, the results remained in remission at 1-year follow up, 50% (6/12) at 2-year follow up and 41.7% (5/12) at 4-year follow up, without continued fecal infusion. Nevertheless, UC patients also have recurrence issues after FMT treatment or other therapies, and some studies reported maintenance FMT administration could sustain clinical efficacy either by capsule or colonoscopy (Sood et al., 2019a; Steube et al., 2019). Long-term maintenance treatment is necessary for some patients.

In one retrospective study regarding the long-term safety of FMT to treat UC, new-onset urticaria, arthritis, depression, and so on, were found in the long-term follow-up, while abdominal discomfort, flatulence, low-grade fever were the most common short-term adverse events (Sood et al., 2019b). Besides, one study reported a serious adverse event of myasthenia gravis in 1 month after FMT treatment (Ding et al., 2019). In our study, all the patients treated with FMT were observed for 4 years, with a maximum of 6.5 years. During the follow-up period, short-term adverse effects were consistent with previous studies by and large, but no long-term adverse reactions caused by FMT were found. Nevertheless, a remarkable adverse event was that one 14-year-old girl underwent furuncle in her legs after the first FMT treatment, though the furuncles scabbed in 1 week without any treatment. The risk of infectious disease transmission for FMT needs to be investigated with more cases collected. On June 13, 2019, the US FDA issued a safety alert regarding two immunocompromised adults who acquired Extended Spectrum Beta-Lactamase (ESBL)-producing E. coli infections following FMT, resulting in one death. Hence, the donor screening protocols for FMT should be further improved by excluding use of stool that tests positive for multi-drug resistant organisms.

Diverse administration routes of FMT have been observed to have similar efficacy in the studies of FMT for the treatment of CDI (Postigo and Kim, 2012; Furuya-Kanamori et al., 2017). We used a combination of colonoscopy and endoscopy as the route of FMT for 11 patients with a clinical response rate 63.6%, and the next 20 patients underwent only a colonoscopy approach, who also experienced good efficacy with a clinical response rate 70%. In the studies of FMT treatment for UC, colonoscopy approach, enema and nasogastric tube are the most common methods of administration. In the five RCT studies, four obtained positive results by conducting under lower GI tract administration (Moayyedi et al., 2015; Paramsothy et al., 2017; Costello et al., 2019), and one obtained negative results by using upper gastrointestinal tract route (Rossen et al., 2015). It is speculated that the upper gastrointestinal administration might be one of the reasons leading to the treatment failure, and FMT via colonoscopy route may be a better choice for UC.

Previous studies mainly focused on mild to moderate UC, including the above-mentioned RCT studies. A few cohort studies have reported the efficacy of FMT on moderate to severe UC (Angelberger et al., 2013; Kump et al., 2013; Ding et al., 2019), and showed only marginally clinical responses. In our study, patients with moderate UC had the highest responses to FMT, followed by, in turn, severe and mild UC. It is worth noting that the FMT treatment was more effective in patients with extensive UC, followed by distal UC. And patients with proctitis had poor responses to FMT, while some patients with extensive and distal UC responded to FMT but were more likely to leave behind lesions in rectum and/or left-sided colon. These findings suggest that lesions confined to distal colon, especially proctitis, had less responsive to FMT and may require more long-term treatment (persistent therapy). It was recently suggested that CD patients with a low microbial load presented a better response to FMT (Sarrabayrouse et al., 2020). Moreover, oral decontamination with antibiotics such as colistin or aminoglycosides has been proposed to enhance the efficacy of FMT (Huttner et al., 2019; Saidani et al., 2019). Thus, an initial low microbial load or reducing the microbial load in UC patients may promote the colonization of donor microbiota thereby enhancing the efficacy of FMT. In fact, the bacterial load various at different location of the GI tract, with a higher bacteria load in the distal colon (∼10¹⁰CFU/ml) than the ileum (∼10⁶CFU/ml), which give rise to different degrees of colonization resistance. Therefore, an intensive FMT regime may be required to efficiently modulate the microbiome for UC patients with the distal colitis.

The gut microbiota profile analysis demonstrated similar results as reported in previous studies of FMT (Costello et al., 2019; Paramsothy et al., 2019). For instance, a lower gut microbiota diversity was observed in the UC patients compared with the healthy donor, which was increased after FMT treatments. The composition of the gut bacterial community was shifted by FMT to a donor-like state and was enhanced after the second administration. These results suggest that repeated FMT treatments enhance the colonization of donor-derived microbiota in UC patients leading to improvement in clinical symptoms. In particular, Faecalibacterium prausnitzii, Bacteroides plebeius, and Bifidobacterium spp. were significantly increased in UC patients receiving FMT, agreeing with some previous studies (Lopez-Siles et al., 2017; Nishino et al., 2018; Lloyd-Price et al., 2019). Interestingly, an increased level of some Bifidobacterium spp. was closely correlated with a reduction in Mayo score, stool frequency, ESR, and neutrophil, indicating a pleiotropic of Bifidobacterium for UC symptoms improvement (Jakubczyk et al., 2020).

As one of the main butyrate producers in the gut, Faecalibacterium prausnitzii is able to induce secretion of anti-inflammatory cytokines, thereby producing energy to the colonocytes and enhancing the intestinal barrier (Sokol et al., 2008; Louis and Flint, 2009; Ferreira-Halder et al., 2017; Lopez-Siles et al., 2017). Hence, the relative abundances of Faecalibacterium prausnitzii increased in UC patients after FMT treatments, suggesting a potential contribution in suppressing inflammation. In addition, we found that a high abundance of Blautia in UC patients at baseline was associated with sustained remission, and an enrichment of Roseburia was particularly associated with therapeutic responses. These results were consistent with previous studies (Paramsothy et al., 2019; Lloyd-Price et al., 2019).

Data about the correlation between Bacteroidetes spp. abundance and UC activity was controversial. In fact, different Bacteroides spp. may have different influences on the disease development in UC patients. For example, Bacteroides plebeius showed a significant therapeutic effect in UC patients from our study, whereas their nearest recognized species Bacteroides vulgatus (Kitahara et al., 2005) was considered as a pathobiont in human gut (Kootte et al., 2017)with a potential of driving injury in the small intestine (Ramanan et al., 2014). From our data, a combination of reduced Bacteroides massiliensis and enriched Bacteroides fragilis seemed to augur a relapse in UC after FMT. Bacteroides fragilis was shown to play a protective role from intestinal inflammation via Toll-like receptor 2 signaling, inducing production of polysaccharide A (PSA) and anti-inflammatory cytokine IL-10 (Mazmanian et al., 2008; Honda and Littman, 2016; Lee et al., 2018). Nevertheless, strains of enterotoxigenic Bacteroides fragilis have functions of tissue invasion and induce severe intestinal inflammation in humans and animals (Zamani et al., 2017). Moreover, a lower level of Bacteroides finegoldii after FMT characterized the patients with sustained remission. In contrast, Paramsothy et al. (2019) found that an increased abundance of Bacteroides finegoldii in donor stool may be associated with observed remission in UC patients receiving FMT. Thus, close species may have significantly distinct characteristics and the same species may contain various strains with different functions. Moreover, our data indicated that Blautia spp. and Ruminococcus bromii were predictors of achieving the sustained remission, which is in line with the findings in Paramsothy et al. (2017, 2019).

Preveotella copri was found to be more abundant in the non-responders at baseline. In a previous study (Paramsothy et al., 2019), P. copri was also found in higher abundance in UC patients who experienced therapeutic failure after FMT. The relative abundance of P. copri presents more stable in the population with IBD than non-IBD populations (Lloyd-Price et al., 2019). It was postulated that Prevotella has an antagonistic relationship with Bacteroides (Ley, 2016). Thus, the response of FMT might be partially impeded by P. copri resisting the colonization of the donor bacteria such as Bacteroides. Moreover, P. copri was suggested to play an immune-modulatory role in human rheumatoid arthritis (RA). About 32% of patients with RA were found to have serum immunoglobulin A (IgA) antibodies specific for P. copri (Larsen, 2017), which was almost absent in healthy subjects. Secretory IgA is the dominant immunoglobulin at the mucosal surface, and it plays a critical role in interacting with the microbiota and maintaining intestinal homeostasis. It was suggested that IgA coating identifies colitogenic bacteria in IBD patients and is associated with treatment outcomes (Palm et al., 2014; Shapiro et al., 2021). Therefore, P. copri-specific IgA may be related to poor response to FMT for UC patients, which deserves further investigation.

This is a series study with only 31 cases at baseline, which were further reduced at the points of F1 (n = 20). Therefore, this small cohort size and samples limited the strength of our conclusions. In addition, there was a loss of endoscopic examination in the long-term follow-up.

Collectively, our study demonstrated that low-intensity single donor FMT treatment is effective and safe for mild to severe UC, and a repeated FMT provides a beneficial efficacy for disease improvement, with a considerable long-term efficacy and safety. The abundances of Faecalibacterium prausnitzii, Bacteroides plebeius and Bifidobacterium spp. increased significantly after FMT treatments in UC patients. The gut microbial composition of UC with different FMT therapeutic outcomes was distinguishable at baseline. In particular, a higher level of Bacteroides fragilis together with lower levels of Bacteroides massiliensis and Roseburia genus might be indicators of relapse given this particular donor. Moreover, an enrichment of Blautia spp. and Ruminococcus bromii together with a reduced Bacteroides finegoldii were associated with long-term remission. Further investigations to identify specific species/strains associated with clinical remission may shed a light for precise treatment, and a donor-recipient matching approach based on the gut microbiota compositions may increase the remission rate in long term.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

The studies involving human participants were reviewed and approved by the Chinese PLA General Hospital Ethics Committee. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin. Written informed consent was obtained from the individual(s), and minor(s)’ legal guardian/next of kin, for the publication of any potentially identifiable images or data included in this article.

YY responsible for the study concept and design. RR analysis and interpretation of the clinical data and drafted the manuscript. XG analysis and interpretation of the microbiome data and critical revision of the manuscript. YS, JL, LP, GS, ZW, BY, and JZ performance of the FMT and revision of the manuscript. All authors had access to the study data and have reviewed and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.