Gut microbiota and allogeneic transplantation

A large number of studies have shown that post-transplantation complications are closely related with the immune system [32‐34]. To clarify the relationship between gut microbiota and allogeneic transplantation, it is very important to discuss the interplay between gut microbiota and the host’s immune system [35, 36] (Fig. 1).

It has been proven that the intestinal immune system can maintain gut bacteria homeostasis and prevent dysbiosis (Fig. 1). Epithelial, mucosal and immune cells at barrier surfaces of the intestinal tract all are important in maintaining gut microbial homeostasis and modulating microbes by producing mucus, antimicrobial peptides or luminal immunoglobulins. Some immune cells are intercalated between intestinal epithelial cells (IECs), such as intraepithelial lymphocytes (IELs), or underneath the epithelium, such as lamina propria immune cells. Others are organized into intestinal lymphoid structures, including isolated lymphoid follicles (ILFs), Peyer’s patches (PPs) and mesenteric lymph nodes (MLNs). Impairment or lack of these immune structures may lead to gut microbial dysbiosis. For example, Gram negative bacteria were over-represented in mice lacking ILFs [37].

Gut microbiota is also important to a host’s immune system. In transplantation, T cells are important in transplant rejection. Interestingly, several studies found that specific gut bacteria species can promote T cell differentiation. In rats, Th17 cell differentiation can be stimulated by Segmented filamentous bacteria (SFB) [38] and Lactobacillus johnsonii [39]. Gut microbiota may also contribute to the generation of memory alloreactive T cells. Hand et al. [40] found that, during a gastrointestinal infection, both the pathogen and intestinal commensal bacteria could cause immune responses and lead to commensal-reactive T-cell memory. Anticommensal T-cell memory may result in a pool of memory cells with cross-reactive T-cell receptors (TCRs). In addition, several gut microbe species have been shown to promote expansion or differentiation of forkhead box protein 3 (Foxp3)-expressing regulatory T cells (Tregs). Some of these colonic Tregs recognize microbial antigens [41, 42]. Additionally, colonic Tregs are increased in germfree mice with a set of defined benign commensals termed altered Schaedler flora [43]. Indigenous Clostridium species have the potential to promote colonic inducible Treg (iTreg) differentiation [44]. Moreover, commensal gut microbiota can also control the development and maturation of mucosal and systemic natural killer T cells (NKTs) [45] and help the development and maturation of lymphoid structures [46].

Collectively, these data indicate that gut microbiota can interact with the immune system. Determining the relationship between gut microbiota and transplant complications, including infections, rejection, GVHD and relapse after transplantation, is urgent.

In recent years, the progress of microbial detection technologies has facilitated studies evaluating the relationship between gut microbiota and allogeneic transplantation. Many animal experiments and human studies have shown that gut microbiota is altered after allogeneic transplantation. When postoperative complications occur, gut microbiota populations and diversity are in a more significant dysbiosis (Table 1).

As discussed above, gut microbial dysbiosis and complications after transplantation coexist. However, the role of a host immune system in the correlation between gut microbial dysbiosis and complications is rarely studied. Similar to pathogens, gut microbes express microbial-associated molecular patterns (MAMPs) such as lipopolysaccharides (LPS) that can be sensed by specialized receptors on various cells, including immune and gut endothelial cells, to communicate with the immune system. There are a variety of pattern recognition receptors (PRRs). The most studied ones are Toll-like receptors (TLRs) and NOD-like receptors [71]. Intracellular adaptors are indispensable in transferring PRR signaling information and MyD88 is the most studied one, which is a molecule downstream of all TLRs except TLR3. After being exposed to MAMPs, PRR signaling in cells can promote the expression of major histocompatibility complex (MHC) and costimulatory molecules, particularly on antigen-presenting cells (APCs) and some endothelial cells [72]. As a result, cytokines such as tumor necrosis factor (TNF), type I interferons (IFNs), interleukin-1 (IL-1) and IL-6 are produced.

The liver can maintain tolerance against harmless antigens derived from commensal bacteria, even if commensal bacteria escape from the gut [73]. However, this surveillance could be temporarily perturbed after liver transplantation. After liver transplantation and during AR, loss of intestinal microvilli, tight junction damage, decrease in fecal secretory IgA and increases in blood bacteremia, endotoxin, and TNF-α were detected, along with dysbiosis of gut microbiota [52]. Furthermore, acute rejection of small intestine allografts was associated with increased TLR expression [74].

There is growing evidence that bacteria and innate PRRs are critically involved in the pathogenesis of acute GVHD after allogeneic stem cell transplantation. In experimental models, reduced GVHD severity, preserved graft-versus-leukemia effects and improved overall survival were found in allo-HSCT recipients that were treated with either anti-endotoxin neutralizing antibodies [75, 76] or an oral LPS inhibitor [77]. Alpha-defensins can selectively kill noncommensal microbes, but preserve commensal ones. However, Eriguchi et al. [70] discovered that Paneth cells were targeted by GVHD, which resulted in an obvious reduction in the expression of alpha-defensins. Moreover, Heimesaat et al. [78] analyzed both gut microbiota composition and impact of bacterial sensing via TLRs in intestinal GVHD (iGVHD). When iGVHD occurred after HSCT, gut microbiota shifted towards Enterobacteria, Enterococci and Bacteroides/Prevotella spp. An analysis of iGVHD in MyD88(−/−), TRIF(−/−), TLR2/4(−/−), and TLR9(−/−) recipient mice showed that bacterial sensing via TLRs was essential for iGVHD development. Increasing numbers of apoptotic cells, proliferating cells, T cells and neutrophils were found within the colons of mice with acute iGVHD. However, compared with wild-type controls, these responses were markedly reduced in MyD88 (−/−), TLR2/4(−/−), TRIF(−/−) and TLR9(−/−) mice. Meanwhile, TLR9(−/−) mice had increased survival rates, whereas TRIF(−/−) and TLR2/4(−/−) mice were not protected from mortality. These results not only emphasize the critical role of gut microbiota, innate immunity and TLR9 in iGVHD but also highlight an anti-TLR9 strategy as a potential novel therapy for iGVHD after HSCT. However, TLR-4 can be activated by MAMPs, especially LPS, and was found to be critical in inducing tissue protective factors and for protection against intestinal cell apoptosis during acute GVHD [79].

To prevent or treat complications and ameliorate the imbalanced gut microbiota after allogeneic transplantation, gut microbial intervention methods are used. Antibiotics, probiotics and prebiotics are most often used. Promising and encouraging results have been obtained (Table 2). However, the role of intestinal decontamination in allogeneic transplantation is still controversial.

Colonization of the intestinal tract by various microbiota precedes infection in many cases, including LT. That lead to the evolution of selective digestive decontamination (SDD), which is initially described by Stoutenbeek et al. [80]. SDD aims to reduce the Gram negative and yeast flora in the gastrointestinal tract using antibiotics and antifungals to prevent infections. An early study [81] suggested that SDD could significantly reduce Gram negative aerobic bacteria and Candida colonization in the gut. It appeared to reduce the high incidence of infection related to these organisms in the early post-transplant period. However, in another study, after performing an SDD with norfloxacin, thirty-two patients had at least one episode of a major bacterial infection. Furthermore, the number of Gram positive microorganisms was greater than that of Gram negative rods and anaerobes [82]. Zwaveling et al. [83] found that SDD did not prevent infections in patients undergoing an elective LT. However, it did affect the type of infection. It demonstrated that Gram positive cocci infections replaced Gram negative bacilli and Candida species infections. Two recent studies showed that the use of SDD prophylaxis in LT patients affected the rate or distribution of infectious complications, duration of hospitalization, antibiotic use, or acquisition of resistant bacteria [84, 85]. However, enhanced ischemia/reperfusion injury assessed by transaminase expression in a mouse model of LT with down-regulated MAMP expression could be partly ameliorated using an antibiotic treatment [57]. Studies also found that GVHD after allogeneic BMT could be ameliorated by eliminating facultative and strict anaerobic microorganisms from the gastrointestinal tract with antimicrobial drugs in the period of time around the allogeneic BMT. Total gastrointestinal decontamination (TGID) was used by Vossen et al. [86] with high doses of non-absorbable antimicrobial drugs while the graft recipient was maintained in strict protective isolation. As a result, a successful TGID of the graft recipient prevented the development of acute GVHD after BMT. In another study, oral administration of polymyxin B could inhibit Escherichia coli outgrowth. Importantly, GVHD after HSCT was ameliorated [70].

Taking probiotics and prebiotics to regulate gut microbiota and reduce the incidence of complications after LT was also reported in several studies. Early enteric nutrition supplemented with a mixture of lactic acid bacteria and fiber reduced bacterial infection rates following LT. Treatment with only fiber led to a low incidence of severe infections [87]. In a BN rat study, Ren et al. [88] found that supplementation with probiotics, including Bifidobacterium and Lactobacillus, and long-term antibiotics promoted partial gut microbial restoration and improved intestinal barrier function in malnourished rats after LT. Similarly, after allograft LT in BN rats, Xie et al. [89] found that the numbers of Lactobacillus and Bifidobacterium in the probiotic group were significantly greater than the antibiotic and allograft groups. Liver injury was significantly reduced in the probiotic group compared with the allograft group. Moreover, the study revealed that probiotics mediated their beneficial effects through an increase of Treg cells and TGF-β and reduction of CD4/CD8 in rats with AR after LT [89].

Probiotic administration is also useful in ameliorating gut microbial dysbiosis after SBT. Compared with non-treated hosts, small bowel histological injuries were significantly ameliorated and BT was reduced in rats with 6 days of probiotics treatment after SBT [90]. Similarly, in mice with an allogeneic stem cell transplantation, modifying the intestinal microbiota using the probiotic microorganism Lactobacillus rhamnosus resulted in a reduced translocation of enteric bacteria to the mesenteric lymph nodes, reduced acute GVHD and improved survival [91].

With the improvement of transplantation techniques and postoperative recovery treatments, allogeneic transplantation has become an effective therapy for a variety of end-stage diseases. However, various postoperative complications, such as infection, rejection and GVHD, still restrict the use of allogeneic transplantation. In recent years, because of the progress in microbial detection technologies, many animal and human studies have shown that populations and diversity of gut microbiota are altered after allogeneic transplantation. When postoperative complications occur, gut microbiota populations and diversity are in a more significant dysbiosis. Furthermore, distinct gut microbial profiles could be potential diagnostic biomarkers of complications after transplantation [52, 61] and even predict a patient’s history before transplantation [64]. However, this needs to be confirmed in more studies. Whether microbial changes cause or follow complications after transplantation is still unclear.

In several studies, intervention methods, such as antibiotics, probiotics and prebiotics, could effectively regulate gut microbiota and reduce the incidence of complications after transplantation. Thus, gut microbiota has the potential to be a novel therapeutic target to restrict, improve and even reverse complications after allogeneic transplantation. However, more studies revealing the definite mechanism of these results are needed. The role of intestinal decontamination in allogeneic transplantation is still controversial. For example, several studies showed that selective perioperative intestinal decontamination did not reduce infectious complications after LT [83‐85]. However, in HSCT, complete intestinal decontamination could significantly reduce the occurrence of postoperative GVHD [86].

Recently, the study of gut microbiota after allogeneic transplantation has significantly progressed. However, there are still many limitations that need to be resolved. Thus far, most studies evaluating gut microbial characteristics after allogeneic transplantation rely on rectal sampling only. However, the diversity and population of microbiota along the gastrointestinal tract are significantly different. Gu et al. [92] found higher phylogenetic diversity in gastric, duodenal, large intestinal and fecal samples than jejunum and ileum samples. Moreover, a greater proportion of anaerobes, such as Bacteroidaceae, Prevotellaceae, Rikenellaceae, Lachnospiraceae, and Ruminococcaceae, were found in the large intestine and feces. However, a larger proportion of Lactobacillaceae were found in the stomach and small intestine. Inferring the status of the whole gut microbiota only by rectal samples may be a challenge. Therefore, the gut microbial profiles along the intestine should be studied in the future. Many studies have been performed on rats. However, there are major differences in the gross anatomy, physiology and food processed in gastrointestinal tract between a mouse and human. Consequently, a host’s gut microbiota may be highly divergent in species, as well. Overall, two major phyla, Bacteroidetes and Firmicutes, account for the dominant gut microbiota in humans and mice [93]. However, in genera taxonomic classifications, 85 % of bacterial genera found in mouse gut microbiota was not present in humans [94]. Thus, to obtain more credible and relevant results, more human studies are needed.

Using metagenomics to investigate fecal samples from 124 European individuals, the MetaHIT consortium found more than one thousand microbial species in the human gut, and over 99 % of them were bacteria [1]. A core healthy human gut microbiome also has been explored and postulated to consist of three enterotypes, typified by the relative dominance of particular groups of organisms: Prevotella, Ruminococcus and Bacteroides spp. [95]. On average, individual microbiota could have long-term stability [96]. However, patients’ gut microbiota pre- and post-surgery [97] or drug consumption [98, 99] may be different. Gut microbiota can also be impacted by many other factors, including host genes [100], immune system [46], geography [101], age [102], weight [103], lifestyle [104], season [105] and diet [106]. For example, significant inter- and intra-individual variations in seasonal stabilities of the human gut microbiota were found [107]. In addition, due to the difference in long-term dietary habits, the human gut microbiome abundance and proportions varied between United States individuals [108]. David et al. [109] also found that short-term consumption of diets composed entirely of animal or plant products altered gut microbial community structure. Most transplant patients are put on special diet during hospitalization, and gut microbiota can be impacted by diet alone. Antibiotics are usually used after allogeneic transplantation, which also can lead to the alteration of gut microbiota [110]. In future studies, these factors should be taken into account.

Current studies showing a relationship between allogeneic transplantation and gut microbiota were mainly concentrated in liver transplantation, small bowel transplantation, kidney transplantation and hematopoietic stem cell transplantation. In addition, the role of a host’s immune system in the correlation of gut microbial dysbiosis and complications is rarely studied and is mostly limited to HSCT [111, 112]. Finding the definite microbiota effect on local and distal immune system that may lead to complications post-transplantation is important. Moreover, molecular pathways by which microbial signals can lead to complications after transplantation should be evaluated in future studies. Modulating infection and alloimmune responses after transplantation may become possible by identifying therapeutic targets. For example, in preclinical and clinical trials, agents that block TLRs are being tested to reduce pathology in septic or autoimmune patients [30]. These should be in-depth studies and extend to other types of transplantation.