The key contribution of culturomics

The first presentation of the postgraduate course was provided by D. Raoult from Marseille, France, who revisited the concept of gut microbiota using culturomics. Studies using 16S rDNA sequencing and metagenomics have opened the field, but these techniques have limitations due to discrepancies that may arise at the level of DNA extraction, sequencing, and bioinformatic analysis, moreover, these techniques are missing minority partners. However, the emergence of the concept of culturomics has enabled the discovery of an important number of new bacterial species, Archae, as well as large viruses, which could not be detected by metagenomic analyses. This approach employed by D. Raoult was initially extremely cumbersome (using 200 different media), but is now more straightforward, using only 17 media, and new bugs continue to be discovered every week.

Gut microbiota and antibiotics

The second talk was also fascinating. M. Blaser (New York, NY, USA) presented the suspected link between gut microbiota disturbances and several chronic diseases for which the aetiology is still doubtful, such as asthma, obesity, diabetes, inflammatory bowel disease, etc. The prevalence of these diseases is increasing worldwide and parallels the increased use of antibiotics. There are now data showing that bacteria that have co-evolved with humans are crucial to their good health. There is an age window when the microbiota is established (0-3 years), and consumption of antibiotics at this age may lead to the disappearance of part of the microbiota and therefore the bacterial diversity which is an important criterium for health. Experiments in mice have shown that antibiotics can modify the composition of gut microbiota leading to increased adiposity and modification of the immune response, favouring several diseases.

With the description of the association between gut dysbiosis and several diseases at present, an interesting approach to consider is how to limit the impact of antibiotics on gut microbiota. A first step is to add probiotics to antibiotic treatments, however, all probiotics are not equal. Saccharomyces boulardii appears to be the leader in this area. All studies have shown a beneficial effect of this yeast on antibiotic-related diarrhoea. Among Lactobacilli, there is one emerging species in this respect; Lactobacillus rhamnosus GG, as revealed by H. Sokol.

Non-probiotic approaches

There are currently non-probiotic approaches to prevent gut dysbiosis which were presented by A. Andremont (Paris, France). Indeed, antibiotics are absorbed in the small intestine and their negative effects on gut microbiota occur essentially in the colon. Thus, first attempts were made to deliver β-lactamase to the colon to avoid the effect of β-lactam antibiotics, and subsequently other alternatives using specifically coated absorbent-like activated charcoal. Experiments in mice and dogs have been successful, especially for fluoroquinolone antibiotics.

Once dysbiosis is established, restoration is possible by faecal microbiota transplantation (FMT). It is possible to successfully treat Clostridium difficile infection using allogenic FMT. Autologous transplantation could be an option in the case of planned antibiotic treatment and would be more acceptable given that the risk of unknown pathogens would be avoided. For FMT, there is a need for common legislation in Europe as well as standardization of the process.

What is already known about this topic?

Infantile colic is a functional gastrointestinal disorder, defined by the Rome criteria. The pathophysiology is still poorly known although a painful intestinal mechanism is suspected. Studies suggest that a disturbance of the gut microbiota, an increase in gut permeability, and lowgrade intestinal inflammation are involved in visceral hypersensitivity. These factors are involved in the pathophysiology of irritable bowel syndrome. In this syndrome, an abnormality of the protease/ antiprotease balance contributes to visceral hypersensitivity and low-grade inflammation.

What are the main results of this study?

The aim of this study was to investigate whether a disturbance of the gut microbiota, associated with an increase in gut proteases, is capable of inducing visceral hypersensitivity. Breastfed infants, aged 1-4 months, were included. There was no difference in pregnancy duration, birth weight, or family history of allergy between the colicky group (n = 7) and control group (n = 7). According to the criteria (Rome III criteria), only the mean duration of crying differed, with 240 ± 95.95 minutes (colicky) versus 24.04 ±19.65 minutes (control).

Rectal faecal transplant enemas from colicky infants induced significant visceral hypersensitivity during rectal distension with larger volumes (+55%, p <0.001 with 0.06 mL; +27%, p <0.001 with 0.08 mL, and +19%, p <0.001 with 0.1 mL) (Figure 1), as well as a better overall response (increase in area under the curve by +33%, p <0.001). In addition, there was a positive correlation between the duration of crying and these distension volumes (Figure 2).

In contrast, there was no difference in protease levels (serine, trypsin-like, and elastase-like) between both groups.

Finally, the analysis of the microbiota showed an increase in Bacteroides vulgatus and Bilophila wadsworthia diversity with an abundance in children with infantile colic. The relative abundance of B. vulgatus was positively associated with mouse hypersensitivity (p = 0.021), but not significantly associated with the duration of crying (p = 0.067).

What are the practical consequences?

This study shows that a component of the stool contents of colicky infants is capable of inducing visceral hypersensitivity in mice, involving either an increase in a nociceptive compound or a decrease in an antinociceptive compound. This is not an impaired proteolytic balance, however this compound might be induced by a different gut microbiota. Further studies are needed to identify this compound and the mechanism of action.

What is already known about the topic?

Obesity and related disorders, such as diabetes, require new therapeutic approaches because the current treatments, including lifestyle changes, and antidiabetic treatments are insufficiently effective in reducing morbidity and mortality. During the last decade, changes in gut microbiota composition have emerged as a potential new therapeutic strategy for improving insulin sensitivity [1]. Several studies have shown that gut microbiota composition is different between thin and obese animals, but also that the microbial composition may reflect impaired metabolic functions, in particular, with a disturbance of ingested food [2]. Finally, these animal studies have suggested a causal link between microbiota abnormalities and metabolic syndrome since the phenotype is transferable by FMT [2]. Although numerous observational studies have suggested correlations between an altered microbiota composition and metabolism in humans, the causality has been difficult to prove. The authors of this study have previously shown in a small pilot study that FMT from thin donors to men with metabolic syndrome induced an improvement in carbohydrate metabolism, together with changes in faecal and duodenal microbiota [3]. Based on these results, the authors have studied the short- and long-term effects of FMT from thin donors on gut microbiota composition in a larger group of men with metabolic syndrome and investigated the pathophysiology of insulin resistance by correlating changes in gut microbiota with several markers of metabolism. In addition, the authors have attempted to identify basal characteristics of the microbiota of recipients in order to explain the improvement in insulin sensitivity in some patients (referred to as metabolic responders) and not in others (non-responders).

What are the main results of this study?

Thirty-eight obese men with metabolic syndrome were included and randomized in the allogeneic (n = 26) or autologous (n = 12) FMT group. FMT was administered by nasoduodenal tube and repeated six weeks later. Eighteen weeks after FMT, no effect was observed on both the microbiota and the parameters of metabolic syndrome. However, six weeks after FMT, the microbiota of the allogeneic FMT group was changed and the metabolic parameters, in particular insulin sensitivity, were improved while no change was observed in the autologous FMT group (Figure 1). In contrast to their previous study, no change in faecal concentration of butyrate was observed [3]. However, allogeneic FMT was associated with an increase in faecal concentration of acetate, as well as changes in the blood level of about 30 metabolites, several of which are involved in the metabolism of tryptophan. In the subgroup of patients who favourably responded to allogeneic FMT, changes were observed in the faecal microbiota, such as, for example, an increase in Akkermansia muciniphila bacterium, and the favourable effects of this bacterium were demonstrated on metabolic syndrome in mice. The authors also highlighted that the basal microbiota composition, as well as a low diversity, were predictive of a good response to FMT.

What are the practical consequences?

Interventions on the gut microbiota, and particularly FMT, are a valid therapeutic approach for metabolic syndrome. Nevertheless, there is high interindividual variability in the response, which may be related to factors from the host, but also from the donor. In addition, the effects are relatively modest with a single FMT and, at best, they are transient. More targeted strategies, such as the use of new generation probiotics (microbiota-derived bacteria) and prolonged administration, are therefore more attractive and are currently being studied.

Antibiotic-associated diarrhoea

The most frequent and best studied consequence of intestinal dysbiosis due to antibiotic intake is antibiotic-associated diarrhoea (AAD). AAD occurs during ±20% of all antibiotic courses and depends on the class of antibiotic, patient risk factors (host factors, hospitalization status, and nosocomial outbreaks), and the definition of AAD. AAD is defined as a change in stool frequency with at least three liquid stools/ day for two consecutive days, occurring during (early onset) or two to six weeks after antibiotic treatment (late onset), when no other cause can be identified (intercurrent viral or bacterial infection, laxative use, or other cause). The class of (broad-spectrum) antibiotics, duration of administration, and age of the patient are risk factors associated with the development of AAD. The administration of some probiotic strains, such as Lactobacillus rhamnosus and Saccharomyces boulardii, reduce the incidence and severity of AAD [4].

Antibiotics early in life

Antibiotics may have a much broader impact, especially when given perinatally or to young infants. Intrapartum antibiotics during Caesarean and vaginal delivery are associated with infant gut microbiota dysbiosis [5]. Dysbiosis acquired perinatally or during early life will induce long-term consequences. Maternal antibiotic treatment (during pregnancy and lactation) results in profound alterations in the composition of the microbiota in mothers and infants [6]. Prenatal antibiotics are associated with a larger body mass index (BMI) at the age of two years [7].

Antibiotics and weight

Sub-therapeutic doses of antibiotics have been used as growth promoters in animal farming since the 1950s [8]. The effect is more pronounced for broad-spectrum antibiotics, and is attenuated when animals are raised in sanitary conditions. Burgeoning empirical evidence suggests that antibiotics also affect human growth. As early as 1955, a randomized controlled trial in Navy recruits showed that a seven-week course of antibiotics led to significantly greater weight gain in the treated group compared with placebo [8].

There is a positive linear relationship between birth weight and BMI in six to seven-year-old children, which is present in both high and low-income countries [9]. The intestinal microbiota of infants is predictive of later BMI and may serve as an early indicator of obesity risk. Bifidobacteria and Streptococci, which are indicators of microbiota maturation in infants, are likely candidates for metabolic programming of infants, and their influence on BMI appears to depend on antibiotic use [10].

Antibiotic exposure before six months of age, or repeatedly during infancy, is associated with increased body mass in healthy children [11]. Repeated exposure to antibiotics early in life, especially β-lactam agents, is associated with increased weight and height [12]. Such effects may play a role in the worldwide childhood obesity epidemic and highlight the importance of judicious use of antibiotics during infancy, favouring narrow-spectrum antibiotics [11]. If causality of obesity can be established in future studies, this will further highlight the need for restrictive antibiotic use [12].

Administration of three or more courses of antibiotics before two years of age is associated with an increased risk of early childhood obesity [13]. In a cohort study, 6.4% children were obese at four years of age [13]. In this cohort, antibiotic exposure was associated with an increased risk of obesity at four years; the more antibiotic courses, the stronger the risk [13]. Children receiving antibiotics in the first year of life were more likely to be overweight later in childhood compared with those who were unexposed (32.4 versus 18.2% at 12 years of age; p=0.002) [14]. Repeated exposure to broad-spectrum antibiotics at ages 0 to 23 months is associated with early childhood obesity [15].

However, some studies have reported contradictory results. Exposure to antibiotics within the first six months of life compared with no exposure was not associated with a statistically significant difference in weight gain up to seven years of age [16].

Gut microbiome - immunity & food allergy

Symbiotic host and microbe interactions are critical for host metabolic and immune development. Early microbiota colonization may influence the occurrence of metabolic and immune diseases [1].

A clear association was found between early-life antibiotic use (three or more courses) and milk allergy, non-milk food allergy, and other allergies in a longitudinal data analysis of 30,060 children [17]. The associations became stronger for younger age and differed according to antibiotic class [17].

Maternal use of antibiotics before and during pregnancy was shown to be associated with an increased risk of allergy to cow’s milk in the offspring, and persisted after adjusting for putative confounders [17]. In children, the risk of allergy to cow’s milk increased with increasing number antibiotics used from birth to diagnosis (test for trend; p<0.001) [18]. Maternal intrapartum antibiotic prophylaxis has been shown to have a significant impact on the infant faecal microbial population, particularly in breastfed infants [19]. Intrapartum antibiotic administration was reported to result in a significant reduction in Bifidobacterium spp. strains [20]. The reduced abundance of these beneficial microorganisms, together with the increased amount of potentially pathogenic bacteria, may suggest these infants are more exposed to gastrointestinal or general health disorders later in life [20].

Antibiotics and the respiratory tract

Antibiotics given during the first week of life is a risk factor for allergic rhinitis and wheezing, while early introduction of solid foods, such as fish, and living on a farm are protective factors for the development of later allergic disease. Antibiotics taken by the infant during the first year of life is associated with an increased risk of asthma [21]. The strength of the association differs with the class of antibiotics, correlating with their effect on the gastrointestinal microbiome [21].

Antibiotic exposure has been associated with increased risk of asthma at three and six years of age [22], in the presence or absence of a lower respiratory tract infection during the first year of life [22]. The adverse effect of antibiotics was particularly strong in children with no family history of asthma (p [interaction] =0.03) [22]. Antibiotic intake was also a risk factor for a positive allergy blood or skin test. According to a systematic review published in 2011, exposure to antibiotics in the first year of life is a significant risk factor to develop asthma. Retrospective studies provided the highest pooled risk estimate for asthma compared with database and prospective studies. Respiratory infections, later asthma onset (asthma at or after two years), and exposure to antibiotics during pregnancy are all independent risk factors.

Antibiotic use in the first year of life is associated with the development of transient wheezing and persistent asthma [23]. A dose-response effect was observed; with five or more antibiotic courses, the risk to develop asthma increased significantly (p<001). There is no association between antibiotic use and late-onset asthma [23]. Antibiotic use in the first year of life is associated with an increased risk of early-onset childhood asthma, starting before three years of age. Reverse causality and protopathic bias may be confounders of this relationship [23].

Antibiotics and IBD

Exposure to antibiotics throughout childhood is associated with IBD, and this relationship decreases with increasing age of exposure to antibiotics. Exposure before one year of age was shown to have the highest risk, decreasing at five and 15 years, although antibiotics at the age of 15 still represented a significant risk factor to develop IBD [24]. Each antibiotic course increased the IBD hazard by 6% (4%-8%) [24]. As with any observational study, causality cannot be inferred and the possibility of data confounded by indication, due to prescription of antibiotics to children with intestinal symptoms of yet undiagnosed CD, should also be considered [25]. Antibiotic use is common in childhood and its potential as an environmental risk factor for IBD warrants scrutiny [25]. Antibiotic exposure has been reported to be significantly associated with Crohn’s disease, particularly in children, but not significantly associated with ulcerative colitis [26].

Antibiotics and diabetes

Exposure to a single antibiotic prescription was not shown to be associated with a higher adjusted risk of diabetes [27], whereas treatment with two to five antibiotic courses was associated with an increase in risk of diabetes for penicillin, cephalosporins, macrolides, and quinolones. The risk increased with the number of antibiotic courses. No association between exposure to anti-virals or anti-fungals and risk of diabetes was demonstrated [27]. Exposure to antibiotics is likely to increase the risk of type 2 diabetes [28]. However, these findings may also represent an increased demand for antibiotics due to an increased risk of infections in patients with yet-undiagnosed diabetes [28]. Antibiotic exposure in childhood is generally not associated with a risk of developing type 1 diabetes [29]. Future studies should investigate the effects of multiple exposures to broad-spectrum antibiotics during the second year of life.

Antibiotics and malignancies

For gastro-intestinal malignancies, the use of penicillin has been shown to be associated with an elevated risk of oesophageal, gastric, and pancreatic cancers [30]. The association increased with the number of antibiotic courses. The risk of lung cancer increased with the use of penicillin, cephalosporins, or macrolides. The risk of prostate cancer increased modestly with the use of penicillin, quinolones, sulphonamides, and tetracyclines. The risk of breast cancer was modestly associated with exposure to sulphonamides. There was no association between the use of anti-virals or anti-fungals and risk of cancer [30].

Study of microbiota and weight status in 935 mother-infant pairs

Microbiota composition and alterations in patients with obesity and new-borns delivered by Caesarean section have been previously evaluated [1]. The Canadian Healthy Infant Longitudinal Development (CHILD) study is a prospective longitudinal birth cohort study which is designed to collect information at time points that are considered to be particularly critical to the health and development of children in terms of defining the influence of genetics, epigenetics, and the microbiome during early life [2]. Hein Tun et al. [1] enrolled 935 mother-infant pairs in the study, and evaluated maternal weight status during pregnancy, infant gut microbiota composition (including 16S ribosomal RNA sequencing) after a median of one month, and body mass index z scores adjusted for age and sex at one and three years of age.

Their results revealed that 7.5% of infants were overweight at the age of one, and 10.4% were overweight at the age of three. Infants born vaginally to overweight or obese mothers were three times more likely to be overweight at the age of one, while Caesarean-delivered infants of overweight mothers had a five-fold risk of being overweight at the age of one. A similar risk was apparent at the age of three. An abundance in Firmicutes species in the infant gut microbiota, particularly Lachnospiraceae, is associated with excessive maternal pre-pregnancy weight and excessive childhood weight at the ages of one and three. The genera of Lachnospiraceae involved differed between infants delivered vaginally and those delivered via Caesarean birth.

Intergenerational transmission of overweight and obesity in case of cesarean delivery

This study of 935 mother and infant pairs revealed evidence of a novel sequential mediator pathway involving birth mode and greater abundance of Lachnospiraceae, regarding the inter-generational transmission of excessive weight and obesity, especially for Caesarean delivery. The prevention of obesity in women of reproductive age is widely recognised to be important both for their health and for that of their offspring. Hanson et al. [3] highlighted that interventions to reduce or prevent obesity before conception and during pregnancy could substantially contribute to achieving the global Sustainable Development Goals, in terms of health, wellbeing, productivity, and equity in current and future generations. Regarding the current progress towards understanding the microbiome, microbiome composition will play an important role in wellbeing in the future.

Over the last 10 years, tremendous advances have been made for cancer patients using new treatment strategies, including immune checkpoint inhibitors that target cytotoxic T-lymphocyte-associated antigen (CTLA-4) and programmed death 1 (PD-1) protein. However, therapeutic responses to these new treatment modalities are often heterogeneous, and some non-responder patients have been reported. The intestinal microbiome has been suggested to be an important host factor for non-responder patients, along with tumour genomics. Previous studies on microbiota and cancer have mainly focused on alterations in the intestinal microbiota of cancer patients (oncobiome) or microbiota precursors in order to define early-stage cancers, mainly colorectal cancers. However, promising new results regarding the influence of intestinal microbiota on anti-tumour immune responses have emerged. Two new studies were published in the first issue of Science this year.

• Gopalakrishnan et al. [2] evaluated the intestinal and oral microbiome in 112 patients with malignant melanoma, receiving anti-PD-1 immunotherapy, and compared baseline microbiota composition between cancer responders and non-responders. They revealed significant differences in the diversity and composition of intestinal microbiota between responders and non-responders. Significantly higher alpha diversity and relative abundance of Ruminococcaceae/Faecalibacterium was observed in responders, and this favourable intestinal microbiota composition has been suggested to enhance systemic and anti-tumour immunity among patients with melanoma. Patients with a low diversity and relatively high abundance of Bacteroidales (unfavourable intestinal microbiome) have impaired anti-tumour immune responses.

• Matson et al. [3] also evaluated the composition of baseline intestinal microbiota in patients with metastatic melanoma before receiving anti-PD-L1 therapy. Among the responders to treatment, Bifidobacterium longum, Collinsella aerofaciens, and Enterococcus faecium have been shown to be predominant members of the microbiota. These authors suggest that the commensal microbiome may exhibit a mechanistic impact on anti-tumour immunity in patients with metastatic melanoma. In the light of the results from these two previous clinical studies, it is thought that baseline intestinal microbiota may play a critical role in mediating the immune- stimulant response in melanoma patients receiving immunotherapy, such as anti-PD-L1 therapy. Further prospective studies are needed to reveal the precise interactions between the microbiome and cancer, not only in melanoma patients, but in terms of potential relevance for all types of cancer and the different treatment strategies.

Antibiotics and gut microbiota

The congress began with a Biocodex symposium on the impact of antibiotics on the gut microbiota. Dr. L. Armand- Lefevre recalled that antibiotics cause major alterations in the microbiota, in particular due to both the broad spectrum of antibiotics as well as high intestinal concentrations. In addition, microbiota resilience following antibiotic therapy may be slow and incomplete. In addition to the well-known short-term side effects, such as diarrhoea, taking antibiotics in early childhood is associated with an increased risk of obesity, allergies, and autoimmune diseases, as specified by Dr. A. Mosca.

How can these risks be reduced?

Firstly, by trying to prescribe fewer antibiotics in a better way, moreover, if antibiotic prescription is necessary, by combining them with a probiotic. This is particularly the case for the probiotic, Saccharomyces boulardii, which limits dysbiosis and facilitates microbiota resilience following the discontinuation of antibiotics. Pr. C. Kelly has also shown that S. boulardii decreases the level of primary bile acids and increases that of secondary bile acids, thus reducing the risk of Clostridium difficile infection.

Our mucus needs fibre to defend us

Our fibre consumption has decreased over recent decades, at least in the West, from more than 150 g per day a few generations ago to a dozen grams per day nowadays. This directly impacts the composition of our intestinal mucus. Pr. M. Desai’s Luxembourg team has shown, using a mouse model, that a low-fibre diet results in the intestinal mucus being more strongly degraded by the microbiota via glycoproteins contained in the mucus as an energy substrate. The resulting degraded mucus no longer plays its role against pathogenic bacteria such as Citrobacter rodentium, resulting in lethal colitis in these mice [1].

New biomarkers in colorectal cancer

The potential role of the microbiota in colorectal carcinogenesis is well known. In a metagenomic study conducted in collaboration with Pr. J. Wang’s team in China, Dr. M. Arumugam demonstrated the existence of a“microbial signature” of colorectal cancer (CRC), based on the identification of four biomarkers which were significantly expressed in CRC patients compared to healthy subjects, in geographically different populations (China, Denmark, France, Austria). Of these biomarkers, two bacterial genes of Fusobacterium nucleatum (Fn) and Parvimonas Micra (Pm) were significantly over-expressed in cases of CRC [2]. Another recent study has confirmed the role of Fn as a biomarker of CRC, which significantly increased the sensitivity of immunological screening and made it possible to retrieve 75% of CRC cases which were negative on immunological testing [3].

With this advance in the recognition of a “microbial signature” of CRC, it may be possible to screen asymptomatic individuals for CRC in the near future based on an immunological test for blood in the stools coupled with microbiota analysis.

Impact of the microbiota on the response to immunotherapy

It has been known for a few years that the gut microbiota has an impact on the efficacy of chemotherapies. Recently, studies have also shown that the microbiota plays a major role in the response to immunotherapy. Based on a study of 26 metastatic melanoma patients, Pr. F. Carbonnel’s team showed that the type of microbiota is correlated with the response to ipilimumab (anti-CTLA-4). Patients with a microbiota rich in Faecalibacterium and other Firmicutes demonstrated a high response rate to ipilimumab and a significantly increased survival rate. The occurrence of ipilimumab-induced colitis was also more common in this group [4]. Similarly, another recent study based on 112 metastatic melanoma patients has shown that their responses to anti-PD-1 varied, and their microbiota, alpha-diversity, and relative abundance in Ruminococcaceae (a family whose main member is Faecalibacterium) were the main predictive factors for response [5].

Focusing on faecal microbiota transplantation

As was the case last year, faecal microbiota transplantation (FMT) was the subject of a workshop and was often referred to in the various presentations. Drs. G. Ianiro and Z. Kassam recalled the very promising results of FMT for ulcerative colitis (two positive randomised controlled trials and one trial which showed a positive trend with FMT although significance was not reached), metabolic syndrome, hepatic encephalopathy, irritable bowel syndrome, and digestive GVH (graft-versus-host) allograft. Apart from recurrent Clostridium difficile infection, repeated FMT appears to be essential for “engraftment” and treatment efficacy. Administration in capsules seems to be the future for this technique but questions remain, in particular around intake dose and frequency, since these parameters are also likely to vary according to indication. Access to FMT is increasingly facilitated by the emergence of “stool banks”, in particular, in countries in which microbiota transplants have been assigned the status of organ/tissue rather than medicine. For example, in the United States, 98% of the population is within a two-hour drive of a centre practicing FMT. Thus, this practice has been widely used in recent years, but remains to be standardised and possibly adapted to patients according to their disease and microbiota.

Akkermansia Muciniphila: a new-generation probiotic?

Discovered in 2004, A. muciniphila is a bacterium that prevails in the mucus. It degrades mucin, stimulates butyrate production, and produces a pili-like protein, Amuc1100, that appears to play an important role in the immune response and barrier function of the intestinal mucus. It appears to have beneficial properties since its presence is inversely correlated to obesity, metabolic syndrome, and some cardiovascular diseases [6, 7]. In mice, its administration has beneficial effects on metabolic syndrome, and the first clinical data in humans should soon be available.

Gut microbiota composition and functions

Dr. Ralijic-Stojanovic stressed that sequence analysis of 16S rRNA has made it possible to confirm that the taxonomy of many cultured intestinal microorganisms was incorrect. For example, Clostridium difficile, which does not belong to the Clostridium butyricum genus, is in fact a distant relative of Clostridium perfringens, in contrast to what was previously assumed [1]. Dr. Ralijic-Stojanovic recalled that between the ages of 7 and 12 years, the GM is still different to that in adults. He concluded by stressing that GM composition is individual, specific, and stable, and may vary depending on age, diet, and lifestyle.

In his presentation, Dr. Bäckhed reported that, while the role of the GM in metabolism is well known (optimisation of caloric availability, intake of enzymes absent in humans, and the role in vitamin K synthesis and short-chain fatty acid production), some more recent publications have revealed that the level of butyrateproducing bacteria is reduced in patients with type 2 diabetes, increased levels of Prevotella improve glucose metabolism, and Christensenellaceae bacteria may be considered as an anti-obesogenic probiotic [2]. Dr. Bäckhed concluded that the GM should be considered as an environmental factor that contributes to host physiology and metabolism.

The GM is very complex, and despite the advances in recent years, all its secrets are yet to be revealed.

Gut microbiota and liver diseases

The role of the GM in liver diseases is being increasingly understood, and some authors have even suggested the existence of a “gut-liver axis”. Dr. Gasbarrini discussed the role of GM in liver inflammation and fibrosis, showing that severe alterations in GM have been observed in cirrhotic patients, with increased levels of Enterobacteriaceae, Veillonellaceae, and Streptococcaceae, and decreased levels of Clostridiacea, Lachnospiraceae, and Eubacteriaceae. Dr. Gasbarrini believes that an insufficient resilience, resulting in adaptation through the acquisition of a dysbiotic microbiota, may contribute to the onset of GM-associated chronic diseases. Intestinal barrier breakdown is the cornerstone to progression of fibrosis and severity of liver cirrhosis.

Another interesting aspect was also discussed: depending on the mechanism underlying liver injury, the GM may induce or prevent hepatic fibrosis. Possible ways to restore a healthy GM include GM modulation (diet, rifaximin, probiotics or prebiotics) or “reinitialisation” through faecal microbiota transplantation.

Dr. Kobyliak presented a poster [3] on a study conducted in patients with non-alcoholic fatty liver disease (NAFLD), who received a probiotic combined with flaxseed oil and wheat germs or a placebo for eight weeks. The results show that the concomitant administration of probiotics and omega 3 reduces liver fat and serum lipid levels, improves the metabolic profile, and reduces the chronic inflammatory state. Dr. Kobyliak concluded that modulating the GM through the use of probiotics is a new option in the management of NAFLD.

This confirms the influence of the GM on liver diseases and the possibility of alternative solutions through the use of probiotics.

Gut microbiota and chronic inflammatory bowel diseases (CIBD)

CIBDs are a heterogeneous group of immune-mediated chronic inflammatory diseases that affect the gastrointestinal tract. There are two main phenotypes of CIBD: ulcerative colitis (UC) and Crohn’s disease (CD). The relationship between the GM and CIBD is the subject of a growing number of publications.

Dr. Sokol discussed the pathogenesis of CIBDs and the fact that they are mediated by activation of the immune system through the GM in sensitive hosts under the influence of the environment. CIBD patients are known to have an abnormal microbiota with reduced diversity, which becomes increasingly reduced when the disease is active. Dr. Sokol stressed that there is an increase in Proteobacteria and a decrease in Firmicutes, which may or may not be correlated with the onset of disease. Thus, the level of adherent/invasive E. coli (Proteobacteria) is significantly increased in CD patients, but not in UC patients or healthy subjects. On the other hand, the level of Faecalibacterium prausnitzii (Firmicutes), which has antiinflammatory effects, is decreased in CIBD patients.

The environmental impact on GM is well known (mode of delivery, diet, antibiotics) and could also affect CIBD. Based on a report of a Danish cohort, Hviid et al. [4] observed a correlation between the number of antibiotic cycles received by a child and the risk of developing CIBD, which is greater for CD than UC.

Regarding pathogenesis, current controversy focuses on whether changes in the GM cause inflammation or vice versa; which came first: the chicken or the egg? Dr. Sokol believes that they are both at the same level, since the clinical manifestations of CIBD occur due to the implementation of a vicious circle between the GM and inflammation, and both may be the cause.

The data provided thus confirm the important role of GM bacteria in the pathogenesis of CIBD; however, our knowledge of the role of fungal GM in the pathogenesis of these diseases is limited. In this regard, Qiu et al. presented a poster [5] setting out the 15 main genera of fungi found in UC patients and healthy subjects (controls). For the Wickerhamomyces, Sterigmatomyces, and Penicillium genera, a positive correlation was observed with the expression of pro-inflammatory cytokines in the colonic mucosa, while the correlation was negative for Nigrospora. The authors concluded that the colonic fungal microbiota of UC patients is different to that of control subjects and that its alterations may be associated with mucosal inflammation and pathogenesis of UC.

Differences between the pathogenesis of CD and that of UC may be explained, in some cases, by the presence of a bacterial or fungal alteration of the GM.

What is already known about this topic?

The main digestive issue in cystic fibrosis (CF) is exocrine pancreatic insufficiency. This is present in 85% of cases and requires supplementation with pancreatic extracts. Despite this supplementation, fat malabsorption may persist. The proportion of Escherichia coli (E. coli) in the human gut microbiome is usually less than 1% but may reach up to 70-80% in patients with CF. There is clonal expansion within a given patient, but the strains are different between patients, which suggests an adaptation of E. coli to its environment. Some strains of E. coli are involved in gut inflammation and colorectal cancer (CRC). According to recent data, digestive inflammation, dysbiosis, or an increased risk of CRC may be associated with CF.

The authors have assumed that a selection of E. coli strains exist in patients with CF, as these strains are capable of surviving in a gut environment containing excess fat and abnormal mucus.

What are the main results of this study?

E. coli was isolated from the stools of six young children with CF and two controls. The authors assessed the growth of these bacteria in a minimal medium for which the only source of carbon was glucose (GluMM) or glycerol (GlyMM) supplementation. E. coli growth was faster in GlyMM plates for strains isolated from CF children compared to those isolated from controls (Figure 1). These differences were not observed under anaerobic conditions. Since the gut environment is essentially anaerobic, oxygen may play an important role, and in the vicinity of the epithelium, the oxygen gradient is minimal.

Genetic analysis has shown that the E. coli strains were distinct. In addition, the eight isolates had more than 11,000 SNPs (single nucleotide polymorphisms) present in one or more children with CF, which were absent in controls. Transcriptomic analysis (RNA-sequence analysis) in the GluMM or GlyMM medium was performed on isolates from two CF children and two controls. These isolates were selected due to their growth in GlyMM medium and their position within the phylogenetic tree. Among the genes expressed differentially, 213 were overexpressed (glycerol absorption and metabolism) and six were under-expressed (glucose cell transport) in GlyMM medium, in a similar way in both CF children and controls. In GluMM medium, only 20 genes were expressed differentially between CF children and controls, compared to 405 genes in GlyMM medium (377 genes were not induced in CF children) (Figure 2). The genes that were under-expressed in CF in GlyMM medium encode proteins involved in stress, acid resistance, and biofilm formation; the genes overexpressed in CF or under-expressed in controls encode proteins involved in growth mechanisms. In CF, the increased growth in GlyMM medium is not related to metabolic reprogramming, but to a loss of growth inhibition and a stress response.

What are the practical consequences?

This study leads to an understanding of the mechanisms involved in CF-associated dysbiosis with regards to E. coli. In order to correct this dysbiosis and limit, at least in part, gut inflammation, it is important to optimise the absorption of intestinal fat to ensure reduced glycerol levels.

An improvement in the intestinal barrier, especially with regards to mucus for this disease, may also reduce the availability of oxygen necessary for the growth of these E. coli strains.

What is already known about this topic?

Cancer-associated microbiota is known to influence cancer development and progression, including CRC. The microbiota of CRC patients is dysbiotic, and several studies conducted without any preconceptions have revealed an enrichment in F. nucleatum in tumour tissues and adenomas compared to non-cancerous colonic tissues [2]. These findings have been confirmed in studies conducted worldwide based on several cohorts of CRC patients. High levels of F. nucleatum have been associated with less T-cell tumour infiltration (T-cell infiltration is, however, a factor of good prognosis) [3], advanced disease, and a lower rate of patient survival. In addition, patients with F. nucleatum- associated CRC often presented with a right colonic location of cancer, BRAF mutation, and microsatellite instability. Studies based on various experimental models have suggested a protumourigenic role of Fusobacterium, which potentiates tumour growth in mouse models of CRC, xenografts derived from CRC cell lines, and in vitro CRC cell lines. The suggested mechanisms range from increased adhesion and invasion of tumour cells to modulation of the host immune response or activation of the Toll-like receptor 4 pathway. However, some animal or cellular studies have shown no carcinogenic effects of Fusobacterium [4].

What are the main results of this study?

To explore the role of Fusobacterium and its associated microbiota in CRC, the authors analysed the microbiota of five independent cohorts of CRC patients. For frozen samples (11 cases), the culture-based detection of Fusobacterium was positive in more than 70% of cases, showing that the bacterium was alive within the tumour tissue. Furthermore, when metastatic tissue was available and of good quality, Fusobacterium was also detected using culture methods. The molecular method (qPCR) demonstrated greater sensitivity, with more than 80% and more than 60% positivity for primary and metastatic tissues, respectively. The Fusobacterium identified in metastatic tissues was the same as that identified in the primary lesion. In addition to Fusobacterium, other bacteria with the same profile were identified, such as Bacteroides fragilis or Bacteroides thetaiotaomicron, however, unlike Fusobacterium, the strains identified in metastases were different from those identified in the primary lesion. In a cohort of 77 patients, the authors observed that there was no link between positive Fusobacterium culture and tumour recurrence. To determine whether the presence of Fusobacterium plays a role in carcinogenesis or is only passively involved in the oncogenic process, the authors used several systems, including human tumour cell xenografts in immunodeficient mice. They observed that Fusobacterium-positive tumours were easily implanted in mice while this was not the case for Fusobacterium-negative tumours. Finally, treatment with metronidazole, a highly active antibiotic targeting Fusobacterium, significantly reduced tumour growth (Figure 1).

What are the practical consequences?

The treatment of metastatic CRC remains a major clinical issue. This study shows that some bacteria of the microbiota, especially those belonging to the Fusobacterium genus, continue to be viable in both the primary tumour and metastatic tissues in most patients, and play a role in CRC progression. The use of antimicrobial treatments targeting these bacteria is therefore a strategy to be considered, while trying to be as specific as possible, because other bacteria may play a protective role or be involved in the response to conventional anticancer therapies and immunotherapies [5].