Diet, intestinal microbial communities and host health

Alan WalkerRowett Institute of Nutrition and Health,

University of Aberdeen

8/9/15

• Human large intestine hosts an enormous number of microbes (“microbiota”)

– 100,000,000,000,000 (1014) bacterial cells

– Greater than the number of human cells

• Thousands of different species colonise

– Most are strict anaerobes

– Each host have a unique and largely stable microbiota

• The cumulative “microbiome” of these cells contains 400x more unique genes than the human genome

– est. 8 million vs ~20-25,000

• Plays a number of key roles in maintaining host health

– Enhances resistance against infection

– Immune system development/maintenance

– Beneficial compound production

– Breakdown dietary fibre

The human intestinal microbiota

Lawley & Walker (2013) Immunology 138, 1-11

Principal substrates available for utilisation by intestinal microbes

Adapted from Cummings & Macfarlane (1991)

Resistant starch

Non-starch polysaccharides

Unabsorbed sugars

Oligosaccharides

Dietary protein

Enzymes / secretions / mucus

5040302010

Amount (gram per day)

Of dietary & intestinal origin: range

Digestibilities for plant cell wall polysaccharides – 7 subjects (Slavin et al J. Nut 1981)

Pure cellulose (Solka Flok) minimal Cellulose (in normal diets) 69.7% (+/-10.7)Hemicellulose 71.7% (+/- 5.4)

[Harry Flint]

• Metabolise dietary components that escape digestion by human enzymes – Endows host with degradative capabilities they have not

needed to evolve themselves

• Vitamin production– K, riboflavin (B2), biotin (B7), folic acid (B9), cobalamin (B12)

• Release of phytochemicals– Phenolic compounds etc

• Primary end products are short chain fatty acids– Acetate (C2), propionate (C3) and butyrate (C4)

• SCFAs are symbiotic compounds– Gut epithelial cells grow on products of bacterial

metabolism

– Derive up to 70% of energy needs from bacterially-produced butyrate

– Increases energy yield from diet (5 to 10% of caloric intake per day)

oligo-, mono-saccharides

complex polysaccharides

Short chain fatty acids

Polysaccharidedegraders

saccharolytic bacteria

Anaerobicfermentation

Fibre utilisation by gut microbes

oxaloacetate

fumarate

succinate

propionyl-CoA

fucose, rhamnose

propane-1,2-diol

propionyl-CoA

DHAP + L-lactaldehydePEP

CO2

lactate

lactoyl-CoA

propionyl-CoA

pyruvate

propionate

hexoses & pentoses

acetyl-CoA

acetoacetyl-CoA

butyryl-CoAacetate

acetyl-CoA

butyrate

acetate

butyryl-P

H2 + CO2

formate

Roseburia inulinivoransRuminococcus obeumSalmonella enterica

methane

Bacteroidetessome Negativicutes

Coprococcus catusMegasphaera elsdenii

Ruminococcusbromii

Eubacterium rectaleEubacterium halliiRoseburia spp.Anaerostipes spp.Coprococcus catusFaecalibacterium prausnitzii

Coprococcus eutactusCoprococcus comes

ethanol

Blautiahydrogenotrophica

Eubacterium halliiAnaerostipes spp.

methanogenicarchaea

hydrogensulphide

sulfate

Desulfovibrio spp.

Major fermentation pathways

Adapted from Louis et al., Nat Rev Microbiol (2014)

• Inhibition of histone deacetylase (butyrate, propionate)

• Altered mucosal gene expression, cell differentiation

• Protection against colorectal cancer, colitis (butyrate)

• Energy source for the colonic epithelium (butyrate)

• Anti-inflammatory effects (including stimulation of Tregs)

• Suppression of colitogenic pathogens (acetate)

• Stimulation of host receptors (FFAR2, FFAR3, GPR109)

• Influence on gut hormones (e.g. GLP-1, PYY) and satiety

• Influences upon gut transit, gut barrier function

• Peripheral energy supply, lipogenesis (acetate)

• Promote intestinal gluconeogenesis (butyrate, propionate)

Impact of gut bacterial short chain fatty acids on the host

Lactate accumulation shown to be due mainly to reduced lactate utilization by other bacteria at pH 5.2 (13C lactate)

0

5

1 0

1 5

2 0

2 5

3 0

C o n tro ls Q u ie s c e n t

c o lit is

M ild c o lit is M o d e ra te

c o lit is

S e v e re

c o lit is

Co

nc

en

tra

tio

n (

mM

)

Lactate accumulation in stool in colitis (Vernia et al 1988)

[Harry Flint]

Belenguer, A. et al (2007) A.E.M. 73, 6526-6533.

• Many diseases are caused by microbes that normally live asymptomatically on the host– e.g. Staphylococcus aureus

(MRSA), Strep throat, gingivitis, acne, meningitis, pneumonia, C. difficile diarrhoea, thrush, UTIs, gastric cancer (Helicobacter pylori).

• A general imbalance (“dysbiosis”) in microbiota composition has been implicated in many disorders– e.g. Inflammatory bowel

diseases, bowel cancer, irritable bowel syndrome, diabetes, liver disease, allergies, atherosclerosis

Round & Mazmanian (2009) Nat. Rev. Immun. 9;313

Host-associated microbes in disease

Anti-oxidant Molecules

Heterocyclic Amines

N-Nitrosamines

Polyamines

Bile Acids

Indoles Anti-Inflammatory Molecules

N

NNH

2NH2

OH

OH

O

OH

H

H

H H

N

NN

NH2

OOH

OH

NN

O

OH

O

OHO

OH

Phytoestrogens

O

OH

Short Chain Fatty Acids

Damaging Protective

OH

O

O

OHOH

H

OH

Impact of microbially derived metabolites on the host

[Sylvia Duncan]

The “Western” diet, microbiota and host health

Koeth R.A. et al. (2013) Nat. Med. 19;576-585

Tang W.H. et al. (2013) NEJM 368; 1575-1584

Walker & Lawley (2013) Pharmacol. Res. 69:75-86.

Altering the intestinal microbiota

Antimicrobials/Pharmabiotics Probiotics Diet Faecal transplants

• The aim of all of these approaches is to shift the composition of the microbiota to a more beneficial state– Is targeted manipulation possible via alterations in host diet?

Short- v long-term impacts of diet on the intestinal microbiota

• Changes in long-term dietary patterns illicit extensive changes in microbiota composition

• Short term dietary regimes can result in reproducible but limited microbiota response

Disappearing human microbiota?• Have host behavioural changes in the urbanisation era introduced changes in

microbiota composition?

Hunter (2012) EMBO reports.13, 498–500

Round & Mazmanian (2009) Nat. Rev. Immun. 9;313

Link between diet, oral microbiota and healthHunter-gatherer Neolithic agriculturist Medieval agriculturist

• Calcified plaque – one of the only preserved records of bacteria

• Densely colonised by oral microbiota

Adler C.J. et al., (2013) Nature Genetics 45, 450-455

Dobney et al., 1994

Image: Dr Jo Buckburry (re-plotted from Moore and Corbett 1978)http://www.leeds.ac.uk/yawya/bioarchaeology/Dental%20disease.html

• Increased caries rate in skeletal records is linked to increased consumption of carbohydrates

• Unknown whether or not these changes were accompanied by shifts in microbiota composition

Oral microbiota changes through history

• Non-pathogenic Ruminococcaceae associated with hunter-gatherers

• Decay-associated Veillonellaceae increase post-farming

Relatively stable diversity estimates

Decrease in bacterial diversity since the industrial revolution(P = < 0.001)

Adler C.J. et al., (2013) Nature Genetics 45, 450-455

[Alan Cooper – University of Adelaide]

Changes in predominant oral pathogens as a result of diet and culture

• These pathogens do not appear to be prominent in hunter-gatherers

• Streptococcus mutans became dominant relatively recently

• Shifts correlate with human behavioural changes (e.g. diet)

• We may need to reconsider what we think of as a normal “healthy” intestinal microbiota

De

Fili

pp

oe

t a

l. P

NA

S(2

01

0)

Age(months)Diarrhoeal cases Healthy controls

PrevotellaEscherichia/Shigella

Pop, M. et al., Genome Biology (2014), 15:R76

Regional variations in gut microbiota composition

[Cooper, P. et al., (2013) PLOS ONE 8,e76573]

Many novel bacteria

Children in :KenyaMaliThe GambiaBangladesh

Links between host diet and microbiota structure

Wu et al. (2011) Science 334;105-8

• Microbiota structures are associated with long term dietary consumption patterns

O’Keefe et al., Nat. Commun. (2014), 6:6342

• O’Keefe et al performed 2-week food exchanges

• African Americans were fed a high-fibre, low-fat African-style diet and rural Africans a high-fat, low-fibre western-style diet

• Resulted in measurable changes in health biomarkers– butyrogenesis, secondary bile acid

synthesis in the African Americans

Links between host diet and microbiota activity

Duncan SH, et al., (2007) A.E.M. 73:1073-8

Impact of low carbohydrate weight loss diets

** P < 0.001

Short chain fatty acids - in faeces

****

**

0

10

20

30

40

50

60

70

80

90

Acetate Propionate Butyrate Iso butyrate Iso Valerate

Faec

al c

once

ntra

tion

(m

M)

1

High

Moderate

Low

Fae

cal S

CFA

le

vels

(m

M)

% Eubacterial (Eub338) count in faeces

***

****

***

*

*

0

5

10

15

20

25

30

35

40

Bac303

Fprau645

Rfla729 +

Rbro

730

Bif164

Erec482

Prop853

Rrec584

Erec -

Rrec

log 1

0 tota

l

Bacterial group (probe)

% E

ub33

8 or

Eub

338

/g

High

Moderate

Low

Mic

rob

iota

co

mp

osi

tio

n

• Low CHO diet leads to reduction in butyrate-producing bacteria

• Reduced faecal butyrate levels

High CHO – 400 g/dModerate CHO – 170 g/dLow CHO – 23 g/d

2

4

6

8

10

12

Ferulic acid 3OMe4OHPPA 34OHPPA 3OHPPA

Maintenance DietLow Carb. Diet 20+/-1 daysLow Carb. Diet 27+/-1 days

P < 0.05 P < 0.001

P < 0.05

Co

nce

ntr

atio

n (

ug

cm-3

)

Ferulic acid derivatives 8 volunteers

Russell WR et al., AJCN (2011) 93, 1062-72

Major fibre derived phenolics in faecal samples

• Low carbohydrate, high protein intake resulted in reduced concentrations of potentially cancer–protective plant phenolic derivatives

14 obese male volunteers with metabolic syndrome (mean age 54 years, mean BMI 39.4 kg/m2)

M NSP RS HPMC

1 wk 3 wks 3 wks 3 wks

Collection of faeces

M NSPRS HPMC

Mean dietary intake [g/d]:

M 427 230 5 28 103 126

NSP 427 138 2 42 102 136

RS 434 275 26 13 109 127

HPMC 201 110 3 22 144 63

CHO: carbohydrate

Diet CHO Starch RS NSP Protein Fat

Weight maintenance

Weight loss

Resistant starch vs non-starch polysaccharide diet

M: Weight maintenance, mixed diet (55% energy from carbohydrates)

NSP : High non-starch polysaccharides (added bran), low RS

RS: High resistant starch (Type III), reduced NSP

HPMC: Reduced calorie intake. Increased % protein, moderate carbohydrate

Walker AW. et al (2011) ISME J 5, 220-230

Volunteer No.

1-P

ears

on

co

rrel

atio

n

Resistant starch vs non-starch polysaccharide diet

Stimulated by NSP Stimulated by RS

• Samples cluster by donor, not by diet

• Sub-group of bacteria are highly responsive to both the NSP and RS-enriched diets– More Ruminococcaceae species

increase with RS

– More Lachnospiraceae species increase with NSP

Salonen, A. et al (2014) ISME J 8, 2218-30.

5040

Resistant starch v non-starch polysaccharide diet

11 12 17 18 19 23 240

20

40

60

80

100

14 15 16 20 22 25 26

volunteer

% r

esid

ual

fae

cal s

tarc

h

RS diet

NSP diet

Residual faecal starch in volunteers100

80

60

40

20

00 10 20 30

Time [h]

% o

f re

sid

ual

sta

rch

V25 faecal sample+ B. adolescentis+ E. rectale+ B. thetaiotaomicron

Volunteer 25 in vitro faecal starch utilisation

+ + R. bromiikeystone species

for starch utilisation

Ruminococcus spp. - qPCR

M NSP

WL

RS

0

10

20

30

40

50

0 10 20 30 40 50 60 70

M NSP RS WL

% o

f u

niv

ersa

l 16

S rR

NA

gen

e co

pie

s

11121718192324

Volunteer

0

10

20

30

40

50

60

70

80

0 10 20 30 40 50 60 70

M RS NSP WL14

1516

20

22

25

26

Volunteer

Time [days] Time [days]

Walker et al. ISME J (2011); Ze et al. ISME J (2012)

Keystone species within the microbiota

Links between diet, microbiota and health

Walker AW & Duncan SH. MNFR (In Prep.)

Conclusions• Long-term dietary patterns have significant impacts on both intestinal

microbiota composition and activity– Fibre consumption appears to be a major driver

• Microbiota composition and activity are strongly correlated with markers of host health/disease– Mechanistic studies now starting to emerge that link diet/microbiota/health

• Specific bacterial groups/species respond strongly to dietary change, but there is inter-individual variation in the groups that respond

• Many gut bacteria appear to be nutritionally specialised; these species are likely to show the greatest responses to dietary manipulation– ‘Keystone’ species may determine the ability to ferment insoluble substrates

• Implications for therapeutic dietary intervention:– The response may depend on the underlying microbiota composition of a

given individual– Need continued and improved characterisation of the microbiota in order to

predict responses to dietary manipulation

Contact : alan.walker@abdn.ac.uk

Protection against

colorectal cancer and

colitis

Exposure to metabolites and bacteria that promote

disease

supply of short chain fatty acidsphytochemicals in colon

colonic pH

Fermentable carbohydrates:

Interplay between dietand microbiota

phenols, amines, indoles, N-nitroso compounds,

H2S, amines, bile acids,faecapentaenes, heme

Nutrient supply to mucosa

Barrier against infection

Release of phytochemicals

[Sylvia Duncan]

Interplay between diet and microbiota on gut health

Pathogen Genomics Group

Julian ParkhillPaul Scott

Bacterial Pathogenesis Group

Trevor LawleyMark Stares

Core Sequencing Teams

Carol ChurcherDavid Harris

Richard Rance

All other members of the Sanger Institute’s sequencing teams

Acknowledgements

U. Maryland

Colin StineMihai Pop

Funding Sources:

St. George’s, University of London

Phil Cooper

University of Adelaide

Alan CooperChristina AdlerLaura WeyrichJohn Kaidonis

Wolfgang HaakCorey BradshawGrant Townsend

University of Warsaw

Arkadiusz Sołtysiak

Johannes Gutenberg University of Mainz

Kurt Alt

Rowett Institute of Nutrition and Health/University of Aberdeen

Microbiology Group

Harry FlintSylvia Duncan

Gillian DonachiePetra Louis

Jennifer InceLucy Webster

Xiaolei ZeAurore BergeratFreda McIntosh

Karen ScottJenny Martin

David Brown (Analytical Chem)

Wendy Russell (Mol Nutr)

Alex Johnstone (OMH)

Gerald Lobley (OMH)

Sylvia Stephen (HNU)

Grietje Holtrop (BioSS)

University of Aberdeen

Keith Dobney

University of Modena and Reggio Emilia

Maddalena RossiAndrea Quartieri