LAUSR

The pathogenesis of primary sclerosing cholangitis (PSC) remains enigmatic; genetic studies gave somehow disillusioning results, diseasemodifying drugs are urgently awaited and therefore progress in PSC therapy is difficult to recognise. Nevertheless, there is continuous and increasing scientific and clinical interest in this clinical conundrum, which is fired by novel omics technologies and innovative animal model techniques. A possible pathogenetic gut– liver axis in patients with PSC suffering from concomitant IBD in 75% (increasingly referred to as PSCIBD to discriminate it from the classic UC or Crohn’s disease) puzzled us for decades and there is increasing genetic and clinical evidence outdating the concept of PSC as an extrahepatic manifestation of IBD. There is accumulating evidence that patients with PSC show complex alterations in the gut microbiome and in addition in the microbiome’s metabolism. 3 In a nutshell, however, most studies reported reduced diversity and shifts in the overall composition of the (faecal) microbiota, but did not point towards specific pathogens and so far did not point towards gamechanging diagnostic or therapeutic avenues. Interestingly, specific antibiotics may have some beneficial effects in patients with PSC and numerous clinical studies are currently ongoing. Bile acid metabolism is modulated by intestinal bacteria and alterations of the gut microbiome, as it is the case with the use of most antibiotics, but interestingly also by sulfasalazine. Consequently, antibiotics and modulation of the gut microbiome affect (1) the bile acid pool size and composition and (2) the liver and gut homeostasis. All together we have the impression something meaningful is currently happening in PSC research improving our understanding of the complex interplay of intestinal bacteria, bile acid metabolism and mucosal immunity, but do we understand to the full? Undoubtedly, gutderived bacteria and their metabolites are currently the hot topic in PSC research. In Gut, Awoniyi et al show that germfree (GF) conditions cause 100% mortality in Mdr2 (Abcb4) mice and that selective pathogenfree faecal microbiota transplantation rescued mice. In addition, treating the mice with vancomycin exaggerated liver injury in parallel with reduction of Lachnospiraceae and increase of Enterococcus and Enterobacteriaceae abundance, together with increased bacterial translocation, presumably from the gut to the liver, of Enterococcus faecalis and Escherichia coli strains. Furthermore, the authors were able to show that Lachnospiraceae and purified SCFA (short chain fatty acids) treatment partially rescued the liver phenotype in antibioticpretreated mice. The PSC Mayo risk scores correlated negatively with Lachnospiraceae and positively with the abundance of E. faecalis and Enterobacteriaceae in a cohort of patients. From this elegant set of data the authors conclude that this may pave the way for personalised targeted therapy in patients with PSC, namely (1) tailored antibiotic treatment, (2) bacteria administration or (3) SCFA supplementation. It is important to note that this is another important paper pointing towards a key role of specific bacteria in PSC after Nakamoto et al elaborated the crucial role of a poreforming Klebsiella pneumoniae in a gnotobiotic mouse model for sclerosing cholangitis (SC). Consequently, the plot thickens, but does it clear up? When interpreting experimental data using Mdr2 mice as PSC model, we continuously have to consider that bile duct injury in these mice is a consequence of free, nonmicellarbound, biliary secreted bile acids due to the lack of biliary phospholipid secretion, as shown already by Smit et al. Consequently, every intervention increasing the bile acid pool size or hydrophobicity of pool composition (eg, cholic acid feeding) will cause aggravation and every intervention reducing bile acid pool size (eg, Asbt inhibitors, FGF19, bile acid binding resins) will ameliorate cholangitis and consequently liver fibrosis. Importantly, it has been known since the 1970s that GF conditions but also antibiotics increase the bile acid pool size in rodents. In this respect, the current findings of Awoniyi et al are therefore in line with previous observations, but the rescue of the liver phenotype with Lachnospiraceae and purified SCFAs is new and fascinating! However, we have to consider that it remains an unresolved question whether biliary bile acids play a role in bile duct injury in PSC, since patients with PSC show regular bile composition in this respect. Therefore, we always have to be cautious when we aim to translate successful Mdr2 treatment to the human situation since this may only work in this specific mouse model. Another interesting finding of the current study is the fact that the PSC Mayo risk scores correlated negatively with the presence of Lachnospiraceae and positively with the abundance of E. faecalis and Enterobacteriaceae, which also raises important questions. Which bacteria are relevant in PSC pathogenesis and could some be used therapeutically in PSC? Where should these bacteria be relevant/active? In the colon, the liver, in bile? How do we explain a potential organotropism of bacteria for large duct injury in PSC when they are gutderived? Is it the bacterium itself, metabolites or the immunoreaction of the host that harms or helps? Again, we have to consider important points when faecal microbiota analysis data sets from patients with PSC are interpreted. Usually such analysis is performed from the stool of patients with PSC and the microbiota of different parts of the colon significantly differ (even more than interindividual). Since PSCIBD is frequently rightsided, microbiota analysis of caecal faeces or mucus may be most interesting for analysis. In addition, disease controls (eg, patients with autoimmune hepatitis or primary biliary cholangitis) should be mandatory, since we recently learnt from Graham et al that ‘gutderivedlymphocytehoming’ may not be specific for PSC, but also occurs in patients with alcoholic liver disease, which may implicate that this finding may be key or a red herring for both diseases. The same may hold true for gut microbiota findings, when disease controls are lacking! Awoniyi et al have to be congratulated for their inspiring paper! The jury is still out whether ‘bugs will be hype or hope’ for patients with PSC.