Development of the Gut-Liver Axis Joseph L. Dempsey, MPH, PhD - - PowerPoint PPT Presentation

A Multi-omic Approach to Understand the Development of the Gut-Liver Axis

Joseph L. Dempsey, MPH, PhD Candidate

Department of Environmental and Occupational Health Sciences

> During the neonatal period, human gut microbiota is high in Lactobacillus, Bifidobacterium, Staphylococcus, and Enterococcus genera (Kundu et al. 2017, Cell). > Adult gut microbiota composition was high in the phylum Firmicutes with increases in Proteobacteria and Bacteroidetes after age 70 (Odamaki et al., 2016, BMC Microbiol). > Knowing the core microbiota during development may help improve chronic diseases associated with the gut microbiota.

Gut microbiota and development

Modified from Kundu et al. (2017) Cell Yatsunenko et al. (2012) Nature

2

Gut microbiota and host diseases

3

> Gut microbiome is associated with health and human disease, but there is limited research on how the developing gut microbiome may contribute to xenobiotic metabolism and disease risk.

Central hypothesis

4

> Age-dependent gut microbiome modulates the metabolome within the gut-liver axis, altering the expression and function of xenobiotic processing genes.

Collect tissues at days:

Conventional (CV) and germ-free (GF) male and female C57BL/6

1 5 10 15 25 60 120 Neonate Young adolescent Adult Multi-omics approach

16S rDNA sequencing of large intestine (Qiime 1.9.1 and 2.0; PICRUSt; FishTaco) Liver transcriptomics (RNA-Seq; STAR aligner v2.5.2b to mm10 gencodeV11 and cufflinks) Targeted metabolomics

• f liver (UPLC-MS/MS;

bile acids and aqueous metabolites) Liver targeted quantitative proteomics and activity-based proteomics (LC-MS)

Experimental Designs

• I. Mice
• II. Humans
• In collaboration with Dr. Sheela Sathyanarayana at Seattle Children’s Hospital, 60 healthy

pediatric human subjects recruited at 4 developmental age groups for fresh fecal microbiome collection  metagenomics and metabolomics

5

> Alpha diversity is highest at Day 1, lowest at early developmental ages, and then increases through adulthood. > For species heterogeneity (β-diversity), there are 4 distinct clusters, separating pre-weaning ages and adults.

Gut microbiota diversity increases through adulthood

6

1000 2000 3000 4000 Faith's Phylogenetic Diversity 2 4 6 8 10 12

Day 5 and Day 10 Day 1 Female Day 15 Day 1 Male Day 25, 60, and 120

Rarefied Sequences α-Diversity β-Diversity

Day 1 Day 10 Day 5 Days 15, 25, 60, and 120

Ontogeny of mouse gut microbiota

7

Age (Days) 1 5 10 15 25 60 120 Age (Days) 1 5 10 15 25 60 120 % Composition 20 40 60 80 100

Male Female

Acinetobacter Lactobacillus

S24-7

Acinetobacter Lactobacillus

f_Lactobacillaceae;g_Lactobacillus f_Bacteroidales S24-7 group;g_uncultured bacterium;s_uncultured bacterium f_Moraxellaceae;g_Acinetobacter f_Bacillaceae;g_Bacillus g_Lachnospiraceae NK4A136 group;s_uncultured bacterium f_Bacteroidales S24-7 group;Ambiguous_taxa;Ambiguous_taxa

• _Clostridiales;f_Lachnospiraceae
• _Clostridiales;f_Lachnospiraceae;g_Lachnospiraceae NK4A136 group
• _Clostridiales;f_Lachnospiraceae;g_uncultured;s_uncultured bacterium

f_Staphylococcaceae;g_Staphylococcus

• _Clostridiales;f_Lachnospiraceae;g_uncultured;s

f_Staphylococcaceae;g_Staphylococcus f_Streptococcaceae;g_Lactococcus

• _Erysipelotrichales;f_Erysipelotrichaceae

g_Oscillibacter;s_uncultured bacterium f_Enterococcaceae;g_Enterococcus

• _Bacteroidales;f_Bacteroidales S24-7 group

g_Ruminiclostridium;s_uncultured bacterium Other Taxa (188)

Age-specific differences in gut microbiota BA metabolism

8

Secondary Bile Acid Synthesis

1 5 1 1 5 2 5 6 2 5000 10000 15000 20000 25000 30000

Secondary Bile Acid Synthesis

D a y 1 D a y 5 D a y 1 D a y 1 5 D a y 2 5 D a y 6 D a y 1 2 5000 10000 15000 20000 25000 30000 35000

Gene counts * * * * * * * * * *

Male Female

* p < 0.05 (one-way ANOVA with

Dunnet’s post hoc)

Wilcoxon Test Statistic 1 2 3 4 5 6 Day 120 Day 1

Bacteria that drive secondary BA synthesis

f_S24-7-OTU276802 f_S24-7-OTU392511 f_S24-7-OTU183446 f_S24-7-OTU400466

• _Clostridiales-OTU387435
• _Clostridiales-348336
• _Clostridiales-OTU768294

f_Lachnospiraceae-OTU352559 f_Ruminococcaceae-OTU338796 f_Lachnospiraceae-OTU352559 f_Ruminococcaceae-OTU338796 f_Ruminococcaceae-OTU190500 f_Enterobacteriaceae-OTU1100639 f_Bacillaceae-OTU1042600 f_Streptococcaceae-OTU516115 Other (2 unique taxa) Other (85 unique taxa)

BAs in the gut-liver axis

Secondary BAs

Modified from Wahlstrӧm et al. (2016) Cell Metabolism 9

CA MCAs CDCA Conjugation with taurine and glycine

T-CA T-CDCA > Bile acids are endogenous detergents and signaling molecules that regulate lipid and glucose homeostasis, energy expenditure, inflammation, bacterial proliferation and gastrointestinal motility. > Gut microbiota produces secondary bile acids from primary bile acids, which are synthesized from cholesterol in the liver.

Primary BAs

10

Top 5 most abundant unconjugated 2° BAs in LI:

1. DCA 2. ωMCA 3. 3α-OH-12keto-LCA 4. 8(14),(5β)-Cholenic Acid-3α,12α-diol 5. LCA

Top 2 most abundant conjugated 1° BAs in LI

(others are minimally present)

• a. T-αMCA
• b. T-CA

BA concentrations are correlated (Pearson) with distinct bacteria in large intestine

Cluster 1

a b

• 1 -0.5 0

0.5 1 Value

4 2 5 1 3

11

BAs and bacteria in large intestine of Cluster 1

nmol/g tissue

0.0 0.2 0.4

Bifidobacterium

0.0 0.5 1.0

Enterorhabdus 0.0 0.5 1.0 Bacteroides 0.0 1.5 3.0 S24-7 (2 taxa) 0.0 0.1 0.2 0.3 Clostridiale Family XIII UCG-001 Lachnoclostridium 0.00 0.02 0.04 Lachnospiraceae

0.00 0.07 0.14 0.21 Intestinimonas 1 2 3 4 Ruminiclostridium 5 0.0 0.2 0.4 0.6

Ruminiclostridium 9 2 4 Erysipelotrichaceae 0.00 0.08 0.16 0.24 Ruminococcus 1 1 2 3

1 5 10 15 25 60 120

0.0 0.4 0.8 Coriobacteriaceae UCG-002 Dorea

1 5 10 15 25 60 120

0.0 0.2 0.4

• L. reuteri

1 5 10 15 25 60 120

0.0 0.2 Akkermansia

1 5 10 15 25 60 120

2 4 6 8

Age (Days)

500 1000 1500

DCA

100 200 80 160 10 20

M F

10 20

LCA

1 5 10 15 120

10 20

T-CA

500 1000

% OTU

T-/MCA MCA 3-OH-12keto- LCA

8(14),(5)-Cholenic Acid- 3,12-diol

Age (Days)

> BA-metabolizing bacteria from Cluster 1 increase with age, corresponding to the increase in secondary BAs.

> The xenobiotic-sensing nuclear receptor PXR modulates the expression of xenobiotic metabolizing genes. > PXR is known to be endogenously activated by the secondary bile acid LCA.

Xenobiotic metabolism and PXR

12 Modified from Kakizaki et al., 2011, Front Biosci Staudinger et al., 2001, Proc Natl Acad Sci USA

Cyp3a

Male Age (Days) 1 5 10 15 120 T-LCA (nmol/g) 0.00 0.05 0.10 0.15 0.20 Female Age (Days) 1 5 10 15 120 Male Age (Days) 1 5 10 15 120 T-DCA (nmol/g) 1 2 3 8 10 Female Age (Days) 1 5 10 15 120 Male Age (Days) 1 5 10 15 120

• MCA (nmol/g)

0.0 0.5 1.0 1.5 6.0 9.0 Female Age (Days) 1 5 10 15 120

* p < 0.05 by enterotype; no gender differences (GLM; Tukey’s HSD)

Age-specificity of secondary BAs between CV and GF mice in liver

* * * * * *

13

CV GF

Expression of important host xenobiotic biotransformation genes in liver is age- and gut microbiome-dependent

14

* p < 0.05 by enterotype

Male Female

Targeted proteomics RNA-seq

Cyp3a11

1 5 10 15 25 60 120 500 1000 1500 2000

Male

1 5 10 15 25 60 120

* * * *

FPKM Relative protein expression (normalized to total protein)

Cyp3a11

1 5 10 15 25 60 120 100 200 300 400 500 1 5 10 15 25 60 120

* * * * *

CV GF

> Phase 1 genes are the top differentially expressed xenobiotic processing genes in both males and females. > Male mice tend to have more differentially expressed than females.

Gender- and age-specific differential expression of xenobiotic processing genes between CV and GF mice

15

1 5 10 15 25 60 120 1 5 10 15 25 60 120 20 40 60 80 100 120

Number of DE DPGs

Phase I Phase II Transporters Antioxidant components Nuclear receptors Bile acid synthesis

Male Female Age (Days) Age (Days)

Activity-based protein profiling (ABPP) to identify functionally active proteins

16 Modified from Sadler and Wright (2015) Curr Opin Chem Biol

> Cyp3a11 RNA expression and ABP abundance are both markedly decreased by lack of gut microbiota.

Fold change between Day 120 male CV and GF mice of detected cytochrome P450s (Cyps) using RNA expression and ABPP

17

RNA expression (FPKM) Activity-based protein abundance (LFQ Intensity)

For RNA expression: * FDR-BH < 0.05 by enterotype For ABP abundance: ± 30%

C y p 1 a 2 C y p 2 7 a 1 C y p 2 a 1 2 C y p 2 a 5 C y p 2 c 2 9 C y p 2 c 4 C y p 2 c 7 C y p 2 d 1 C y p 2 d 2 6 C y p 2 d 9 C y p 2 f 2 C y p 2 j 5 C y p 3 a 1 1 C y p 3 a 1 3 C y p 3 a 2 5 C y p 4 a 1 C y p 4 a 1 2 a C y p 4 a 1 4 C y p 4 f 1 4 Relative LOG2 fold change (GF/CV)

• 4
• 3
• 2
• 1

1 2

* * * * * *

> RNA expression and ABP abundance are both decreased by lack of gut microbiota for Gsta2, Gsta4, and Gstp1, and increased for Gstt2.

Fold change between Day 120 male CV and GF mice of detected glutathione-S-transferases (Gsts) using RNA expression and ABPP

18

Gsta2 Gsta3 Gsta4 Gstk1 Gstm1 Gstm2 Gstm3 Gstm5 Gstm6 Gstm7 Gsto1 Gstp1 Gstt1 Gstt2 Gstt3 Gstz1 Relative LOG2 fold change (GF/CV)

• 2.0
• 1.5
• 1.0
• 0.5

0.0 0.5 1.0

* * * * *

RNA expression (FPKM) Activity-based protein abundance (LFQ Intensity)

For RNA expression: * FDR-BH < 0.05 by enterotype For ABP abundance: ± 30%

> There is an age- and gender-divergent regulation of gut microbiota in mice that in part determines the metabolome. > The gut microbiome contributes to the regulation of the transcriptome, which can also modulate the activity and detoxification capacity of xenobiotic biotransformation genes.

Summary of the developmental regulation of the transcriptome and metabolome of the gut-liver axis

19

How does the gut microbiome and metabolome of fecal samples from human children compare?

Scope of developing human fecal microbiome study

20

7-24 months 5-12 years 2-5 years 1 week to 6 months 16S rDNA sequencing n=5 n=5 n=5 n=5 Aqueous metabolites n=10 n=17 n=16 n=14

Characterization of age-specific human fecal microbiome

21

1 w k

• 6

m t h s 7

• 2

4 m t h s 3

• 5

y r s 5

• 1

2 y r s 2 1

• 3

y r s % Composition 20 40 60 80 100

• _Bacteroidales;f_Bacteroidaceae;g_Bacteroides
• _Clostridiales;f_Ruminococcaceae;g_Faecalibacterium
• _Bifidobacteriales;f_Bifidobacteriaceae;g_Bifidobacterium
• _Clostridiales;f_Lachnospiraceae;g_Blautia
• _Enterobacteriales;f_Enterobacteriaceae;g_Escherichia-Shigella
• _Clostridiales;f_Lachnospiraceae;g_Fusicatenibacter
• _Clostridiales;f_Lachnospiraceae;g_
• _Clostridiales;f_Clostridiaceae 1;g_Clostridium sensu stricto 1
• _Clostridiales;f_Lachnospiraceae;g_Roseburia
• _Bacteroidales;f_Rikenellaceae;g_Alistipes
• _Clostridiales;f_Ruminococcaceae;g_Subdoligranulum
• _Clostridiales;f_Ruminococcaceae;g_Ruminococcus
• _Clostridiales;f_Lachnospiraceae;g_Anaerostipes
• _Lactobacillales;f_Enterococcaceae;g_Enterococcus
• _Clostridiales;f_Lachnospiraceae;g_Lachnoclostridium
• _Clostridiales;f_Lachnospiraceae;g_[Eubacterium] rectale group

Other Taxa (155)

n=5 per group n=1

Characterization of age-specific human fecal metabolites

22

Valine Norvaline Shikimic acid Phenylpyruvic acid Phenylalanine 4-Aminobutyric acid Carnitine Creatine Proline Betaine Leucic acid Citraconic acid Creatinine 4-Methyl-2-oxopentanoic acid/Ketoleucine/Ketoisoleucine Alanine Ornithine Glycine Sarcosine Taurine Phosphocreatine Tryptophan Indole-3-lactic acid Glutaric acid Anthranilic acid Normetanephrine Pipecolinic acid Phenylacetic acid Glyceric acid 3-Aminobutyric acid

metabolites

• 1.0
• 0.5

0.0 0.5 1.0

value

z score

1 week – 6 months 7 – 24 months 3 – 5 years 5 – 12 years * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

#

pregnenolone sulfate Methylmalonic acid 3-Phenyllactic acid Phthalic acid p-Tolylacetic acid/4-Ethylbenzoic acid Nicotinic acid 4-Pyridoxic acid 4-Imidazoleacetic acid 2,3-Dihydroxybenzoic acid N-Acetylmuramic Acid

metabolites

• 1.0
• 0.5

0.0 0.5 1.0

value

1 week – 6 months 7 – 24 months 3 – 5 years 5 – 12 years

z score

* * * * * * * * * * * * * * * * * * * * * * *

Amino acids and amino acid metabolism Drugs, vitamins, and bacterial cell wall

> Understanding the basal interactions of the developing gut-liver axis begins to identify critical time windows of liver development and signaling that may be harnessed for therapeutic potential (e.g. synthetic bile acids).

Summary of the developmental regulation of the transcriptome and metabolome of the gut-liver axis

23

PI: Dr. Julia Cui, PhD, DABT Cui Lab Moumita Dutta, PhD Joe Lim Mallory Little David Scoville, PhD Kris Weigel Angela Zhang Undergraduate students Former lab members Cindy Li, PhD Sunny Cheng, PhD Yasmin Everson Guensli Siginir

Acknowledgments

24

Prasad Lab Bhagwat Prasad, PhD Deepak K. Bhatt, PhD Arizona Metabolomics Laboratory Haiwei Gu, PhD Northwest Metabolomics Research Center (NW-MRC) Daniel Raftery, PhD Dongfang Wang UW Gnotobiotic Animal Core Jisun Paik, PhD, RD, Director Charlie Hsu, VMD, PhD Olesya Pershutkina Doug Miller Lynn Hajjar, PhD DEOHS IT Brian High John Yocum Seattle Children’s Hospital Sheela Sathyanarayana, MD, MPH Trina Colburn, PhD Mark Abbey-Lambertz Pacific Northwest National Laboratory Aaron Wright, PhD Funding from NIH and UW NIH: T32 ES007032-39, GM111381, ES019487, and ES025708 UW: Start-up Funds from Center for Exposures, Diseases, Genomics, and Environment (P30 ES007033) and the Sheldon Murphy Endowment.