Autoimmune liver disease and microbiome

Question 1

what are the different autoimmune liver and bile disorders?

Answer 1

primary biliary cholangitis (PBC)
autoimmune hepatitis (AH)
primary sclerosing cholangitis (PSC)
IgG4-related cholangitis (IC)

Liver biopsy and blood profile – hard to distinguish between the diseases, need further analysis
- most people with PSC also have IBD

Question 2

what is PSC?

Answer 2

Bile ducts drain bile that liver makes – tubular, usually smooth

In PSC, immune system attacks liver and bile ducts – scar tissue and narrowing of bile duct – bile backflow – infection, scarring of liver irreversible (cirrhosis), bile gets stuck
- No therapy slows disease progression
- Affects 20s-40s
- 40-50% end up needing liver transplant - no cure
- 70-80% have IBD
- Immunosupression works for bowel component, not liver component

Question 3

what is the difference between PSC and ulcerative colitis?

Answer 3

Ulcerative colitis – 1% of pop have this in UK; Inflammation of the left hand side

In PSC, inflammation on right side

Question 4

what is the prevalence of PSC in IBD patients?

Answer 4

proportion of people with PSC get IBD: 70-80%
- 80% of people with PSC get inflammation in colon

proportion of people with IBD get PSC: 5% of people who have widespread inflammation of bowel will have features of PSC
- 1 in 5 people with IBD have PSC (8-21% get radiological features of PSC)

Question 5

why is the gut microbiome important for host physiology?

Answer 5

Many gut bacteria in bowel
Close association between gut and liver inflammation, partly due to microbiome
- Healthy microbiome should be rich and diverse = important for healthy function – transforming bile acids, maintaining gut barrier, vitamin metabolism and absorption, preventing outgrowth of pathogens from food
- Autoimmunity risk is reduced with more diverse microbiome e.g. in India

Question 6

what happens to the microbiome in autoimmunity?

Answer 6

Key features of chronic immunity is loss of diversity
- Pathogens can outgrow – pathobiont overgrows in bowel and not kept in control by microbiome due to lack of diversity
- Short chain fatty acids are reduced

Question 7

what determines change to the microbiome in health and disease?

Answer 7

Genetics don’t affect gut microbiome – few autoimmune conditions have genetic drive (except celiac)
- Main contribution is via environmental factors and co-habitation – these determine bacteria and their functions in microbiome

Question 8

does the microbiome change occur before or after chronic liver disease?

Answer 8

Longitudinal microbiome capture to see where microbiome changes before or after disease – is it causative?
- Realisitcally, this can’t be done from birth and for whole life

In IBD – can do colonoscopy every year to sample bowel – gives insights:
- Reduction in diversity is an early feature of IBD = associated with amine release that activates adhesion molecule for recruitment to liver
- Reduction in short chain fatty acids and FXR activation
- Drop in barrier function is late

Question 9

why do microbiomes differ across patients?

Answer 9

Bowel from patients with PSC will be different due to heterogeneity
- Due to co-variants e.g. use of antibiotics, diet

Question 10

what is dysbiosis?

Answer 10

disturbance in the pool of bacteria of the gut

Question 11

what consistent changes to the microbiome are seen in PSC and PSC-IBD?

Answer 11

Consistent changes
- Upregulation of enterococci
- IBD only is different to IBD-PSC enterotype

Question 12

do PSC-IBD mechanisms differ to IBD alone?

Answer 12

Samples of colonic biopsy of IBD-PSC vs ulcerative colitis patient
- proetomics, transcriptomics
- Look at map of net function in bowel

Certain pathways more affected in IBD-PSC than UC
- Particularly in bile acid handling – bile homeostasis disturbed more than other pathways
- bile acid-mediated inflammation

Question 13

how is lymphocyte trafficking affected in PSC-IBD?

Answer 13

In PSC
- Endothelial vessels, where lymphocytes enter, express certain adhesion molecules on their surface
- These are usually only seen in bowel
- Liver starts to express a gut-like phenotype that recruits gut immune cells – bowel postal code
- MADCAM1, CCL25 only found in bowel but found in liver in PSC – recruits bowel immune cells from portal vein to ligand (a4b7 for MADCAM1, ccr9 for ccl25)

Question 14

what is VAP1?

Answer 14

hepatic vascular adhesion protein
- VAP1 is an adhesion molecule and enzyme that breaks down amine

Question 15

how is VAP1 affected in autoimmune liver disease compared to other diseases and in health?

Answer 15

When we look at explanted tissue, there is suggestion that VAP-1 is overexpressed in PSC relative to normal liver, and other immune-mediated liver diseases. s
-More amine substrate in autoimmunity is provided to VAP, which leads to more VAP activity

Normal liver, AIH, PBC, PSC
- More VAP1 in PSC liver – increased amine load to activate

Question 16

what are the consequences of hepatic VAP1 activation?

Answer 16

VAP1
- Give amines = produces aldehyde (drives fibrosis), ammonia, hydrogen peroxide
- hydrogen peroxide leads to upregulation of MADCAM1
- This means immune cells of IBD patients accumulate in the bowel
- But in PSC, they also go to liver due to increased amine load activating VAP1 which activates MADCAM1

Question 17

how does cysteamine exposure affect mice?

Answer 17

Cysteamine exposure can lead to colitis in mice - induces colitis
- cysteamine is often employed in the experimental induction of colitis in mice., and can be generated by a gut epithelial enzyme known as Vanin-1
- Colitis patients have bacteria which produce cysteamine: Inflamed colonic epithelium and Escherichia and Enterobacter spp. – main provider of cysteamine
- Overexposure to cysteamine can lead to colitis and cancer
- Stop cysteamine exposure/inhibit cysteamine generation in mice: can stop colitis phenotype and protect from CRC

Question 17

is VAP1 activity substrate-dependent? what is the most important substrate?

Answer 17

yes
- it was identified that cysteamine was the substrate associated with the greatest enzyme efficiency…
Amine substrate is cysteamine is the most potent VAP1 activator

Question 18

how is vanin-1 implicated in mice colitis?

Answer 18

Vanin-1 overexpression has been shown to induce colitis through increased cysteamine production; furthermore, deletion of vanin-1 abrogates colonic inflammation – a feature abolished through exogenous cysteamine provision.

Question 19

does VAP1 inhibition have any effects on PSC?

Answer 19

Can we inhibit VAP1? Anti-VAP1 agent:
- Not big change in liver biochemistry
- Some patients responded, others didn’t
- This is because liver diseases are heterogeneous, early and late stages
- Targeting with one agent no matter the stage isn’t effective

liver disease stages should inform targets

Question 20

how do vitamin B-producing bacteria correlate with PSC?

Answer 20

Down regulation of vit. B6 synthesis is associated with poorer clinical outcomes in PSC
- Species of bacteria that produce vit B6 are low in IBD-PSC patients compared to healthy control
- Vit B6 levels in blood = transplant free survival is reduced
- Normal vit B6 = improved survival

Question 21

what are the possible therapeutic avenues for PSC-IBD?

Answer 21

Gut microbial depletion to prevent upregulation of enterococcus
- Target with vancomycin oral antibiotic
- Phase 2a trial with vancomycin and then off therapy to see colonic and liver outcomes
- used in PSC with active colitis

Gut microbial replacement:
- Improve change gut microbiome – faecal microbiota transplant
- used in PSC-IBD without advanced fibrosis

Reduced gut toxin absorption:
- Carbalive – carbon nanoparticle beads which stick to endotoxins and sequester them
- Byproducts of gut bacteria can’t translocate into liver
- used in PSC-IBD with advanced fibrosis

Question 22

how does facel microbiotia transplantation from healthy donor vs PSC patients affect mice?

Answer 22

In animal model:
- germ-free mouse (no gut microbiome, sterile) – give faecal transplant (stool from healthy human and give to mouse) – recolonise bowel – nothing happens
- If stool is from PSC patient – mouse develops histoloical features of PSC
- Stool can induce disease in germ-free mouse – blood vessels and bile ducts have concentric fibrosis of PSC

Question 23

which specific bacterial species are associated with PSC/UC?

Answer 23

• K. pneumoniae
• P. mirabilis
• E. gallinarum
• enterococcus

Question 24

what was the PSC-IBD vancomycin open-label clinical trial?

Answer 24

- Over 16 years of age, with requirement for active inflammation in bowel - moderately active PSC-IBD - baseline: Many patients had active inflammation in gut despite current IBD therapy - 4 weeks on treatment, 4 weeks off

Question 25

how was IBD activity affected in the vancomycin trial?

Answer 25

Calprotectin – marker of stool inflammation - Baseline calprotectin were elevated (should be below 200) - Treatment: they fell - Post-treatment: most relapsed Antibiotic clearly treats something in the bowel to reduce inflammation

Question 26

how were clinical and endoscopic outcomes affected in the vancomycin trial?

Answer 26

Symptoms: - Partial mayo score – IBD activity reduced with treatment then relapsed after treatment stopped - Endoscopy – treatment reduced IBD activity and colitis

Question 27

how was liver biochemistry affected in the vancomycin trial?

Answer 27

Liver parameters, e.g. bilirubin, fell on treatment and then relapsed

Question 28

how was alpha diversity affected in the vancomycin trial?

Answer 28

Microbial diversity needs to be enriched for IBD-PSC patients, but antibiotics reduce diversity - Need to see overgrowth of beneficial players – this wasn’t seen - Alpha diversity – measure of microbiome richness – this shrunk with treatment: narrowed diversity at intra-patient level Oral vancomycin is associated with significant reduction of alpha diversity (richness and evenness within a single individual) This diversity reduction ‘paradoxically’ correlates with a drop in faecal calprotectin

Question 29

how was beta diversity affected in the vancomycin trial?

Answer 29

Beta diversity – measure richness level between individuals - Treatment also reduced this

Question 30

how were taxonomic changes affected in the vancomycin trial?

Answer 30

Beneficial bacteria which release short chain fatty acids were reduced Downregulation of enterococcus Upregulation of pathological players upon treatment e.g. Escherichia - Despite this, colitis got better?? how?

Question 31

overall, how does vancomycin affect the microbiota?

Answer 31

Oral vancomycin is associated with PSC-IBD remission and improvement in liver enzymes - But oral vancomycin depletes your gut bacteria away from ‘healthy’ immunoregulatory SCFA-producing taxa towards a ‘more dysbiotic’ pro-inflammatory phenotype. Treatment depletes beneficial bacteria, but the patients were improved - why?

Question 32

how does colitis remission correlate with bile salt hydrolase (BSH)?

Answer 32

Bile salt hydrolase (BSH)-producing bacteria are knocked out with vancomycin treatment, e.g. K/O of bile acid transformation - BSH levels correlate strongly with faecal calprotectin - BSH dropped quickly – associated with calprotectin: Higher BSH, higher calprotectin - Vancomycin reduces BSH is BSH producer depletion cause of remission or just an effect of vancomycin?

Question 33

were resistant enterococci observed with vancomycin treatment?

Answer 33

Vancomycin could cause outgrowth of antibiotic-resistant bacteria (vancomycin-resistant enterococci (VRE)) -- But resistant enterococcus strain outgrowth didn’t occur in the trial - There was upregulation of pediococcus, but these aren’t pathogenic, they are benign

Question 34

what do secondary bile acids promote? what does this indicate about vancomycin?

Answer 34

Seocndary bile acids promote differentiation of effector T cells to Treg phenotype whilst inhibiting TH17 pathway k/o of secondary bile acid bacteria with vancomycin is bad

Question 35

how does oral vancomycin affect secondary bile acids?

Answer 35

Oral vancomycin is associated with a reduction in secondary bile acids Despite this, patients were getting better confusion

Question 36

how does oral vancomycin affect mouse model of PSC?

Answer 36

In a PSC mouse model, biliary fibrosis is exacerbated with antibiotics and germ-free conditions: - Without antibiotics = fibrosis - With vancomycin = worse bile duct fibrosis = worsened disease - More staining around vessels = bad - Giving vancomyin in mice can make conditions worse in mice: more bile acid concentrations, bile duct barrier dysfunction, fatal liver injury, reduced FXR signalling But patients feel better?

Question 37

summary of PSC-vancomycin:

Answer 37

Short duration oral vancomycin is associated with clinical remission of PSC-IBD Vancomycin leads to depletion of colonic mucosal microbial alpha diversity Shift in colonic mucosal taxa paradoxically appears to be away from immunoregulatory short chain producing Clostridia species and towards a proinflammatory inducing phenotype - associated with functions of microbiome Major changes in inferred microbial metabolic pathways: depletion of bile salt deconjugating bacteria possibly associated with improvement in PSC-IBD activity Phase Iib / phase III trial planning underway - trial needed to compare control subsets to actively treated

Question 38

what are the outcomes of vancomyin for PSC patients?

Answer 38

Oral vancomycin is associated with improved IBD outcomes in PSC - Vancomycin compared to non-treated in clinic - 65% children with colitis retained remission from IBD - Validates the trial

Question 39

has faecal microbiota transplant been studied much?

Answer 39

Faecal microbiota transplantation (FMT) in u. colitis - Trials in IBD space, less for liver

Question 40

has FMT been successful?

Answer 40

Remission in 28% patients in the donor FMT groups compared with 9% patients in the placebo groups 1/3-1/4 can attain remission with fecal transplant compared to 9% in control groups People with established IBD therapy without remission could obtain remission with FMT

Question 41

what is the RESTORE-UC trial?

Answer 41

Randomised control trial for allogeneic FMT: - Screen vigorous donor questionnaires for selection of healthy donors for the trial Looked at autologous stool as control vs allogeneic from healthy volunteer - 4 once weekly treatments - Endpoint was steroid free clinical remission

Question 42

what were the outcomes of the RESTORE-UC trial?

Answer 42

Steroid-free clinical remission is associated with lower baseline inflammatory burden, lower bacterial cell count and higher prevalence of Bacteroides 2 - No statistically significant difference in proportion of people who responded in allogeneic vs autologous - 3.9% more in autologous obtained primary endpoint - Trial stopped Enterotype transition (change on treatment) didn’t change much in healthy donor FMT - Microbiome didn’t change - But did change in specific subsets of patients

Question 43

what attributes can determine response to FMT?

Answer 43

Higher the mayo score (degree of IBD activity) = less likely to respond Active but mild-moderate disease = more response Fewer no. bacterial cells in recipient = better response Loose, moist stool = better response

Question 44

does FMT work in mice?

Answer 44

Faecal microbiota transplantation attenuates biliary injury in an experimental model of sclerosing cholangitis - Mice with bad biliary disease given healthy donor stool improves bile/liver disease

Question 45

can FMT be used for PSC?

Answer 45

only a small pilot study, with 10 PSC patients - primary outcome - safety - 9 with IBD and 9 with larger duct PSC - Single inoculation of healthy donor FMT in colonoscopy - Safe - Reduction in 5/9 patients, but one inoculation isn't enough

Question 46

what is the FARGO trial?

Answer 46

FARGO trial for PSC patients treated with FMT~: - Receive 8 FMT via routine colonoscopy each year, topped up by enema - Multiple treatments for engraftment of donor stool

Question 47

why may using healthy donors for FMT be limited?

Answer 47

May not be easy just to use healthy donor FMT, as it may not work in recipient - May need to think like blood matching – need donor and recipient to match for clinical outcome