Influence of feeding and management on gut microbiota and barrier function

Influence of feeding and management on gut microbiota and barrier function Dept. Animal Science Aarhus University

Mechanisms by which diet and feed additives may modulate the intestinal ecosystem and barrier function

Microbiological toolbox Stomach (Sto) Liver Caecum (Cae) Pancreas Small intestine (Si) Rectum (Rec) Colon (Co) Sampling Digesta/ Digesta mucosa Microbiological Toolbox DNA Extraction & Purification Pig gastrointestinal tract (GI-tract) Pig Intestine Direct analysis DNA analysis Metabolite analysis T-RFLP Community fingerprints 1. Raw DNA Fragmentation Adapter ligation Strand separation Pyrosequencing 2. Fragments Bead attachment (1:1) Droplet capturing (1:1) PCR amplification Selective culturing 4. Sequencing Deposited in microwells (1:1) 3. Clones on beads Deposited in microwells (1:1) 3

Density of selected groups of cultivable bacteria in various section of the gastrointestinal tract of slaughter pigs Log CFU/g digesta 11 10 9 8 7 6 Total anaerobe Bacteroides Lactobacilli Bifidobacteria Coliform bacteria Enterococci 5 Yeast 4 Sto SI1 SI2 SI3 SI4 Cae Co1 Co2 Rec Gut segment

Microbial activity in various regions of the gastrointestinal tract of pigs 30 g ATP/ g digesta 25 20 15 10 5 Low fiber Oat bran Wheat bran 0 Sto SI 1 SI 2 SI 3 Cae Co1 Co2 Co3 Co4 Gut segment 5

Characterise the culturable bacteria in the gastrointestinal tract of pigs Are we able to cultivate gut bacteria?

AARHUS Relative distribution of isolates on species Eschericia coli Prevotella sp1 Strep. alactolyticus S. hyointestinalis L. acidophilus johnsonii Lactobacillus sp2 Caten. mitsuokai Coprococcus sp Citrobacter sp. Mitsuokella mitsuokai Megasphaera sp1 Acidamin. fermentans Clostridium perfringens Fusobacterium sp12 Bacteroides vulgatus Fusobacterium sp11 F. mortiferum Eubacterium spa Selenomonas sp Peptococcus sp3 Megasphaera elsdenii Staph. warneri Known species Unknown species Not determined 0 5 9 species Small Intestine (580 isolates): 19 known species (99%) 0 unknown species (0%) 5 not (yet) determined (1%) Total 24 species 0 10 20 30 40 50 60 % of isolates

AARHUS Relative distribution of isolates on species Eschericia coli Prevotella sp1 Strep. alactolyticus S. hyointestinalis L. acidophilus johnsonii Lactobacillus sp2 Caten. mitsuokai Coprococcus sp Citrobacter sp. Mitsuokella mitsuokai Megasphaera sp1 Acidamin. fermentans Clostridium perfringens Fusobacterium sp12 Bacteroides vulgatus Fusobacterium sp11 F. mortiferum Eubacterium spa Selenomonas sp Peptococcus sp3 Megasphaera elsdenii Staph. warneri Known species Unknown species Not determined Colon (553 isolates): 27 known species (47%) 42 unknown species (49%) 21 not (yet) determined (4%) Total 90 species % of isolates

16S AARHUS rdna match of predominant colon isolates Proportion of isolates Isolate/species (%) Prevotella sp1 16.8 Streptococcus alactolyticus 8.6 Streptococcus hyointestinalis 6.3 Escherichia coli 5.5 Acidaminococcus fermentans 4.6 Megasphaera sp1 3.9 Lactobacillus acid. johnsonii 3.6 Selenomonas sp1 3.6 Mitsuokella multiacidus 2.9 Prevotella sp22 2.6 58.4 Similarity GenBank 16S rdna match (%) OTU 16 98,8 Streptococcus alactolyticus 100 Streptococcus hyointestinalis 99.8 Escherichia coli 99.8 Acidaminococcus fermentans 99.4 Megasphaera elsdenii 96.0 Lactobacillus johnsonii 98.7 URB 4C28d-14 98.1 Mitsuokella multiacidus 98.7 OTU 18 97.2 OTU: Uncultured Bacterium (Pig) URB: Uncultured Rumen Bacterium

Comparison between isolates and most dominant clones* Clones (9270) Isolates** (791) Rank Closest related species (similarity) % of clones % of isolates 1 Streptococcus alactolyticus (99.7) 7.9 22.8 2 Escherichia coli (100.0) 4.8 12.2 3 Lactobacillus sorbius (99.2) 4.5 12.0 4 Faecalibacteria prausnitzii (92.6) 4.0 <0.1 5 Lactobacillus johnsonii (99.8) 3.2 5.3 6 Faecalibacteria prausnitzii (97.6) 2.9 <0.1 7 Sporobacter termitidis (90.7) 2.4 8 Gemella haemolysans (87.1) 2.1 9 Lactobacillus reuterii (99.5) 1.9 4.7 10 Eubacterium rectale (92.6) 1.6 * Data from Leser et al. 2002, ** Similarity to clones >97%

Bacteroides-Prevotella cluster Bacteroides fragilis DSM2151 T' B83 B97 Bacteroides uniformis JCM5828 T Bacteroides stercosis ATCC43183 T B81 Bacteroides vulgatus ATCC8482 T otu-29 B220 otu-28 Prevotella zoogleoformans ATCC33285 T B84 UB adhufec55 Bacteroides merdae ATCC43184 T B86 Bacteroides distasonis DSM20701 T' B175 otu-34 otu-233 B89 Bacteroides splanchnicus NCTC10825 T Vr96 Rp84 otu-22 Rp15 Prevotella ruminicola subsp. brevis ATCC19188 T' Prevotella ruminicola subsp. ruminicola ATCC19189 T B116 otu-15 Cr62 Ls7 Ls16 B173 Vr15 Vr40 otu-9 B95 B123 Ls8 Rp58 otu-112 Prevotella bryantii DSM11371 T' B4 Ls10 B87 B112 otu-16 Prevotella oulora ATCC43324 T Prevotella veroralis ATCC33779 T B91 otu-18 Prevotella ruminicola TC2-3 (rumen isolate) Cp65 Cr25 Cr29K2 otu-17 otu-344 Wc11 Ls17 0.03 Isolates Type strains (Strains) Uncultured bacteria clones Leser et al. 2002 (otu) Suau et al. 1999 (adhufec) (Algorithm: Neighbor-Joining)

Prevotella ruminicola cluster Vr96 Rp84 otu-22 Rp15 Prevotella ruminicola subsp. brevis ATCC19188 T' B116 otu-15 Ls7 Ls16 B173 Vr40 Vr15 otu-9 B95 B123 Ls8 Rp58 otu-112 0.03 Prevotella ruminicola subsp. ruminicola ATCC19189 T Cr62 B4 Prevotella bryantii DSM11371 T' Ls10 B87 B112 otu-16 Prevotella oulora ATCC43324 T Prevotella veroralis ATCC33779 T Isolates Type strains (Strains) Uncultured bacteria clones Leser et al. 2002 (otu) Suau et al. 1999 (adhufec) (Algorithm: Neighbor-Joining)

Conclusion Defined >120 species from 2200 isolates > 60% no match to any known species Despite high diversity, 75% of the isolates were covered by eighteen OTUs (<1% of isolates each) Good agreement between our isolates and OTUs obtained from pig clone library however some dominant pig clone library OTUs were not between our isolates and some of our isolates could not be related to any of the OTUs in the pig clone library Most pig gastrointestinal bacteria are cultivable

UNIVERSITET Aarhus, 3. October 2011 Local Strain Collection of Gastrointestinal Bacteria 127 Species or Operational Taxonomic Units (OTU) 49 matches already known species 56 apparently new species 22 yet to be analyzed or lost 15

Abudance in most dominating bacteria genera in various sections of the gastrointestinal tract of pigs using pyrosequensing (Looft et al., 2014) Ileum Caecum % abudance Colon Faeces

Age related change in bacterial populations in faeces of pigs Prevotella Anaerobacter Lactobacillus % abudance Streptococcus Faecalibacterium Escherichia Age of pigs (weeks) Kim et al., 2011

Important differences between the microbial ecosystem in the stomach/small intestine and the caecum/colon Stomach and small intestine Few bacterial species (20-30) Low density/high activity Compete with the host for digestible components (use around 6% of the energy) Produce toxic compounds Prevent growth of pathogenic bacteria Caecum and colon Many bacterial species (>400) High density/low activity Support the host with energy (around 16%) Produce toxic compounds May support growth of pathogenic bacteria

Factors that modulate the microbiota in the gastrointestinal tract Feed Diet composition Feed processing Fermented liquid feed Prebiotics Low protein diets Additives In-feed antibiotics zinc oxide organic acids probiotics Plant extracts/species

Feed additives Results from Danish weaner trails Kjeldsen, 2008

AARHUS Effect of Virginiamycin on the microbial activity in various regions of the gastrointestinal tract of pigs g ATP/ kg metabolic weight 350 300 250 200 150 100 50 Control 350 300 250 200 150 100 50 50 ppm Virginiamycin 0 Sto SI 1 SI 2 SI 3 Cae Co1 Co2 Co3 Co4 Gut segment 0 Sto SI 1 SI 2 SI 3 Cae Co1 Co2 Co3 Co4 (Adapted from Jensen, 1988)

Effect of Virginiamycin (20 ppm) and Spiramycin (20 ppm)on cadaverin concentration in various regions of the gastrointestinal tract of piglets Cadaverine concentration (umoln/g DM) 12 10 8 6 4 2 0 Control Virginiamycin Spiramycin Stomach SI 1 SI 3 SI 5 Cae Col Gut segment Adapted from Dierick et al., 1985

Effect of in-feed antibiotics on abundance of most dominating bacteria in ileal lumen samples % abundance *100 ppm chloreteracyclin, 100 ppm sulfamethazine, 50 ppm penicillin) Looft et al., 2014

Effect of in-feed antibiotics on most dominating bacteria in ileal mucosa samples Non-medicated % abudance Looft et al., 2014

Microbial activity in various regions of the gastrointestinal tract of pigs 30 g ATP/ g digesta 25 20 15 10 5 0 Low fiber Oat bran Wheat bran Sto SI 1 SI 2 SI 3 Cae Co1 Co2 Co3 Co4

Relative distribution (diets and segments) Acidaminococcus fermentans 20 15 Isolates (%) 10 5 0 Ileum Caecum Colon Low fiber Oat bran Wheat bran

Effect of feed structure on Salmonella prevalence Salmonella prevalence (% positive pigs) 15 10 5 0 a b Fine/pelleted Coarse/meal Jørgensen et al., (1999)

Effect AARHUSof grinding and pelleting on ph, organic acid concentration and the density of coliform bacteria in stomach content from slaughter pigs ph 4,6 4,4 4,2 4,0 3,8 3,6 3,4 3,2 Organic acid concentration (mmol/kg) 40,0 5,5 Stomach Stomach Stomach a a a b 30,0 20,0 10,0 0,0 b ab a a F/M G/M F/P G/P F/M G/M F/P G/P F/M G/M F/P Coliform bacteria (log CFU/g content) 5,0 4,5 4,0 a ab b ab G/P

Bacteria genera (T-RFs) in the stomach of pigs. Number of pigs in which the bacteria were found (out of 6) bp Tentative identifiction Fine Coarse 206 Unknown 3 5 223 L. delbreuckii subsp delbreueckii 0 6 253 L. delbreuckii subsp bulgaricus 0 4 267 L. mucosae 0 6 390 M. jalaludinii/multiacidia 0 6 406 L. reuteri 0 6 574 Unknown 0 6 581 S. alactolyticus/hyointestinalis 2 4 589 S. ruminantum 0 5 591 M. elsdenii 0 5 595 L. sorbius 2 5 611 Weisella sp. 0 6 605 L. curvatus 0 5 Data from Mikkelsen et al., 2005 and Canibe et al., 2005

Lawsonia posetive pigs (%) 60 50 40 30 20 10 0 Effekt af feedstructur and tylosin on infection with Lawsonia intracellularis Feed structure Johansen et al., (2003) 240 pigs per group 43 Fine/pelleted P = 0.06 31 Coarse/meal Lawsonia posetive pigs (%) 60 50 40 30 20 10 0 Tylosin Pedersen et al., (2003) 185 pigs per group 52 Control P = 0.04 39 Tylosin (20 ppm)

Effect of particle composition (fine vs. coarse barley) on growth performance and feed conversion. Growth performance (g/day) 925 900 875 850 825 800 100 50 25 15 0 Feed conversion (Fus per kg growth) 3,2 3,0 2,8 2,6 2,4 2,2 2,0 100 50 25 15 0 % Coarse barley Jørgensen et al., 2002

The stomach as a barrier Increased microbial activity in the stomach Increased production of organic acids decrease ph Kill enterobacteria entering the stomach Lower proliferation of enterobacteria in the small intestine Fewer enterobacteria excreted with faeces Breaking the vicious circle where pigs in problematic herds infect each other and are re-infected through consumption of faeces

Three ways to reduce ph and increase the concentrations in stomach content: Addition of organic acids to the feed Use coarse grounded feed Use of fermented liquid feed

Specific growth rate (min-1) 0,00-0,02-0,04-0,06-0,08-0,10-0,12-0,14-0,16 AARHUS Effect of various organic acids (100 mm) on specific growth rate of coliform bacteria in gastric content at ph 4.5-168 Control -60-216 -36-30 -6-18 Formic Propionic Butyric Lactic Benzoic Fumaric -0,18

Specific growth rate (min-1) 0,04 0,03 0,02 0,01 0,00-0,01-0,02 AARHUS Effect of various organic acids (100 mm) on specific growth rate of coliform bacteria in content from the small intestine at ph 5.5 36 Control -312 Formic 582 Propionic -660 Butyric 60 Lactic -60 Benzoic 30 Fumaric

Effect of organic acid on diarrhea in a herd with ETEC AARHUS (Weaning at 25 days, 4 weeks in experiment) problems a b bc cd d df e N= 4, one replicate is one pen with 12 pigs. Antibiotics: 40ppm lincomycin og 44ppm spectinomycin Tsiloyiannis et al., 2001

Effect AARHUS of 0.7% formic acid /0.7% lactic acid and 2% benzoic acid on growth performance and feed utilisation by piglets after weaning (4-10 weeks) Performance index: Growth performance (g/day) 400 380 360 340 320 300 Feed utilization (FUs/kg growth) Control FA/LA BA Control FA/LA BA 2 1,9 1,8 1,7 1,6 1,5 Control: 100 (a) LA/FA: 111 (b) BA: 121 (c)

Effect of organic acids and zinc oxide on diarrhea and killing out range of weanears *: 1% lactic acid 1% formic acid og 0,5% benzoic acid **: 2500 zinc as zinc oxide

Definitions Liquid feed (LF): water and feed are mixed shortly before feeding. Fermented liquid feed (FLF): feed and water are mixed and soaked during a certain period of time, at a certain temperature before feeding.

Characteristics of dry feed, LF and FLF DF LF FLF DM% 89.3 27.3 24.6 ph - 5.9 4.4 Lactic acid bact. (log CFU/g) <4.3 7.2 9.4 Enterobacteria (log CFU/g) <4.7 6.2 <3.2 Energy (MJ/kg DM) 18.7 19.0 19.0 Lysine (g/16 g N) 6.0 5.8 4.8 LMW-sugars (% DM) 3.6 2.9 0.1 Canibe and Jensen, 2003

Microbial populations in the GIT 10 Lactic acid bacteria 7 Coliform bacteria a b c a a b a a b 9 a ab b a a b 5 a a b 8 Sto SI3 Cae Co2 3 Sto SI3 Cae Co2 DF NFLF FLF

Dry feed as a risk factor for Salmonella incidence * * * *** ** study 1 study 2 study 3 study 4 study 5 Bager, 1993; Bager & Emborg, 1994; Dahl, 1997, Stege et al., 1997; Dahl (unpubl.)

Effect of FLF on swine dysentery Positive culture from faeces of B. hyodysenteriae Clinical disease after infection with B. hyodysenteriae 60 a a % o f to tal d ay s 40 20 b b 0 DF FLF Lindecrona et al. (2000)

Durations of faecal excretion (days) 17.5 15.0 12.5 10.0 7.5 5.0 2.5 0.0 Effect of FLF on infection with Lawsonia intracellularis * STD (n=19) FLF (n=22) FAC (n=23) LAC (n=16) COARSE (n=18) Diet

AARHUS Effect of zinc and copper on daily weight gain in piglets after weaning (at 4 weeks) Daily weight gain (g/day) 300 280 260 240 220 200 180 160 140 120 Day 0 to 14 after weaning Day 14 to 35 after weaning b ab a a Daily weight gain (g/day) 750 700 650 600 550 500 450 bc c b a 100 400 100 ppm Zn 0 ppm Cu 100 ppm Zn 175 ppm Cu 2500 ppm Zn 0 ppm Cu 2500 ppm Zn 175 ppm Cu Poulsen et al. (2003)

Bacterial activity Bacterial activity (ATP content of digesta) ATP (µg/g wet digesta) 30 25 20 15 10 100 ppm Zn 0 ppm Cu 100 ppm Zn 175 ppm Cu 2500 ppm Zn 0 ppm Cu 2500 ppm Zn 175 ppm Cu ** * ** ** ** } Low ZnO 5 ** * } High ZnO 0 Sto Si1 Si2 Si3 Cae Co1 Co2 Co3 Segment Levels of significance: * P Zn < 0.05 and ** P Zn < 0.001 (Højberg et al. 2005, AEM 71:2267)

Lactic acid bacteria (MRS agar) Bacterial counts (log CFU per g digesta) 10 9,5 9 8,5 8 7,5 7 P Zn < 0.0001 P Cu < 0.01 P Zn < 0.0001 P Cu = 0.38 P Zn < 0.0001 P Cu = 0.42 P Zn < 0.01 P Cu = 0.86 6,5 Stomach Ileum Caecum Colon 100 ppm Zn 0 ppm Cu 100 ppm Zn 175 ppm Cu 2500 ppm Zn 0 ppm Cu 2500 ppm Zn 175 ppm Cu

T-RFLP (caecum) Low ZnO High ZnO (Højberg et al. 2005, AEM 71:2267)

Probiotica Viable defined microorganism, which alter the microbiota in a way that exert beneficial effects on the host The three most commonly used organisms as probiotics: bacillus, yeast and lactic acid producing bacteria (lactobacillus, bifidobacterium and enterococcus) Inconstancy in results properly doe to: dosage and type of strains, environment and diet type Needed: strains as well of doses that gives consistently good results Mechanism not completely understood, a few have been proposed: inhibit pathogen adhesion by steric hindrance or competitive exclusion production of products with antibacterial activity (bacteriocins or OA) modulation of the host immune system

Development of the population of Enterococci in faeces from piglets fed a control diet or a diet supplemented with 10 9 Enterococcus faecium 8,0 Enterococcus in feed Log CFU/ g faeces 7,0 6,0 5,0 Control Enterococcus 4,0 1 3 5 8 10 12 15 17 19 22 Days after weaning

Effect of NDO on yeast 6,0 5,0 Control FOS GOS b b c b b b b 4,0 ab a a a a 3,0 Stomach Distal SI Caecum Mid colon Mikkelsen et al., 2002

Boar taint Skatole Produced in the large intestine from microbial metabolism of tryptophan Can be reduced by dietary intervention (high fermentable fibre, protein type) Androstenone Produced in the testicles From there to blood, saliva, and from liver to (bile), urine, addipose tissue. 60

Task 2.6 Experiment with paraxites/chicory root Two-factorial experiment(n=72) Pigs slaughtered at three times (after 5, 9 and 12 weeks on the experimental diets) -Chicory root - worm Control (n=3x6) + worm Worm (n=3x6) + Chicory root (25%) hickorye (n=3x6) Chickory + worm (n=3x6) Nudelworm

Indols and p-cresol in colon content 62

Androstenone and skatole in back fat Control Control Chicory root Chicory root Worm Worm Male Female Male Female Male Female Male Female Androstenone (ng/g) 1,48 0,01 1,21 0,00 1,24 0.01 1,53 0,01 Skatol e (µg/g) 108,7 57,0 93,8 70,1 0,0 0,0 0,0 0,8 Indole (µg/g) 21,4 8,6 14,6 15,2 7,3 1,8 7,4 6,1 63

Low protein diets Source and level known to affect enteric health of weaned pigs Source: animal protein better than plant protein more knowladge on microbial composition and activity needed Level: low protein diets supplemented with synthetic AA prevent post weaning diarrhoea without compromising growth performance

Conclusion Feed FLF: Increase gut health but reduce growth performance may be a way to make probiotics cost effective. Feed structure: Increase gut health but reduce growth performance Prebiotics: Inconsistent results. More research needed. May not be cost effective. Additives ZnO: increase growth performance and reduce PWD, but gives environmental problems OA: right doses and types have beneficial effects on growth performance and health Probiotics: Inconsistent results. More strains should be tested. May not be cost effective