© 2016 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
A reference gene catalogue of the pig gut microbiome
LiangXiao1†,JordiEstellé2†,PiaKiilerich3†,YuliaxisRamayo-Caldas2,ZhongkuiXia1,Qiang
Feng1‡,SuishaLiang1,AnniØyanPedersen4,NielsJørgenKjeldsen4,ChuanLiu1,5,Emmanuelle
Maguin6,JoëlDoré6,7,NicolasPons7,EmmanuelleLeChatelier7,EdiPrifti7,§,JunhuaLi1,8,
HuijueJia1,XinLiu1,XunXu1,StanislavD.Ehrlich7,9,LiseMadsen1,3,10,KarstenKristiansen1,3*,
ClaireRogel-Gaillard2*andJunWang1,3*
Supplementary information
SUPPLEMENTARY INFORMATIONARTICLE NUMBER: 16161 | DOI: 10.1038/NMICROBIOL.2016.161
NATURE MICROBIOLOGY | www.nature.com/naturemicrobiology 1
Supplementary Figure 1 | Distribution of microbial gene (a) and MGS (b)
annotated against the NR database. Only annotated genes/MGS (green areas in
panels a & b) were included in the analysis. 50.8% of the NR genes and 69.1% of the
MGS could be annotated to the Bacteria super-kingdom, demonstrating that many
unknown genes are detected in the pig gut microbiota. Further taxonomic annotation
was carried out for the NR genes assigned to Bacteria and Archae, with 41.2 and 7.6%
genes annotated at the phylum and genus levels, respectively. Less than 1% of the
genes could be annotated at the species level (0.33%).
Genus
Prevotella Bacteroides Clostridium Ruminococcus Eubacterium Butyrivibrio Lactobacillus Other Unknown
Species
Escherichia coli
Methanobrevibacter smithii
Faecalibacterium prausnitzii
Ruminococcus obeum
Phascolarctobacterium succinatutens Prevotella copri
Acidaminococcus fermentans Dorea longicatena
Lactobacillus amylovorus Other
Unknown
Superkingdom(50.8%)
Phylum(41.2%)
Genera(7.6%)
Species(0.33%)
3,903,5853,782,287 3,170,194689,465587,128444,626 25,336
561,792
Superkingdom
Bacteria Archaea UnknownFirmicutes Bacteroidetes
Proteobacteria Euryarchaeota
Fibrobacteres Unknown
Prevotella Eubacterium Lactobacillus Dorea Clostridium Butyrivibrio
Ruminococcus Methanobrevibacter Helicobacter Coprococcus Catenibacterium AcidaminococcusPhascolarctobacteriumStreptococcus
Unknown
Escherichia coli
Unknown
Phylum Genera Species(69.1%)
497222
(49.1%)
353144
(4.6%)
3312 132
(0.14%)
Gene Annotation
MGS Annotation
a
b
Superkingdom Unknown Phylum Unknown Unknown Unknown
BacteriaArchaea
Eukaryota
Bacteroidetes Spirochaetes Euryarchaeota Verrucomicrobia Unknown
Firmicutes Proteobacteria Actinobacteria Fibrobacteres Other
Superkingdom Unknown Phylum Unknown Genus Unknown Species Unknown
Supplementary Figure 1
Supplementary Figure 2 | KEGG (a) and eggNOG (b) annotation of the NR
genes assigned to bacterial species (left hand) and annotation of the NR genes
clustered into MGS (right hand). The functional annotation based on all NR genes
and on the genes clustered in the MGS was found to be consistent. COG: Cluster of
Orthologous Groups.
P : Inorganic ion transport and metabolism
a
b
Supplementary Figure 2
Fu
nc
tio
n c
las
s
Number of matched genes
Cellular Processes −− Cell CommunicationCellular Processes −− Cell Growth and Death
Cellular Processes −− Cell MotilityCellular Processes −− Transport and Catabolism
Environmental Information Processing −− Membrane TransportEnvironmental Information Processing −− Signal Transduction
Environmental Information Processing −− Signaling Molecules and InteractionGenetic Information Processing −− Folding, Sorting and Degradation
Genetic Information Processing −− Replication and RepairGenetic Information Processing −− Transcription
Genetic Information Processing −− TranslationHuman Diseases −− Cancers
Human Diseases −− Cardiovascular DiseasesHuman Diseases −− Immune System Diseases
Human Diseases −− Infectious DiseasesHuman Diseases −− Metabolic Diseases
Human Diseases −− Neurodegenerative DiseasesMetabolism −− Amino Acid Metabolism
Metabolism −− Biosynthesis of Other Secondary MetabolitesMetabolism −− Carbohydrate Metabolism
Metabolism −− Energy MetabolismMetabolism −− Enzyme Families
Metabolism −− Glycan Biosynthesis and MetabolismMetabolism −− Lipid Metabolism
Metabolism −− Metabolism of Cofactors and VitaminsMetabolism −− Metabolism of Other Amino Acids
Metabolism −− Metabolism of Terpenoids and PolyketidesMetabolism −− Nucleotide Metabolism
Metabolism −− Xenobiotics Biodegradation and MetabolismOrganismal Systems −− Circulatory System
Organismal Systems −− DevelopmentOrganismal Systems −− Digestive System
Organismal Systems −− Endocrine SystemOrganismal Systems −− Environmental Adaptation
Organismal Systems −− Excretory SystemOrganismal Systems −− Immune SystemOrganismal Systems −− Nervous SystemOrganismal Systems −− Sensory System
0e+00 1e+05 2e+05 3e+05 4e+05 5e+05
Distribution of KOs from MGS
Fu
nc
tio
n c
las
s
Number of matched genes0e+001e+052e+053e+054e+055e+05
Distribution of KOs from annotated species
Distribution of COGs from MGSDistribution of COGs from annotated species
Fu
ncti
on
cla
ss
Number of matched genes
F : Nucleotide transport and metabolismS : Function unknown
T : Signal transduction mechanismsN : Cell motility
K : TranscriptionY : Nuclear structure
E : Amino acid transport and metabolismV : Defense mechanisms
Z : Cytoskeleton− : Unknown
Q : Secondary metabolites biosynthesis, transport and catabolismM : Cell wall/membrane/envelope biogenesis
C : Energy production and conversionL : Replication, recombination and repair
A : RNA processing and modificationO : Posttranslational modification, protein turnover, chaperones
J : Translation, ribosomal structure and biogenesisW : Extracellular structures
B : Chromatin structure and dynamicsH : Coenzyme transport and metabolism
D : Cell cycle control, cell division, chromosome partitioningR : General function prediction only
I : Lipid transport and metabolismG : Carbohydrate transport and metabolism
U : Intracellular trafficking, secretion, and vesicular transport
0e+00 1e+05 2e+05 3e+05 4e+05
Fu
ncti
on
cla
ss
Number of matched genes0e+001e+052e+053e+054e+05
Supplementary Figure 3 | KEGG annotation of gut microbiome biological
functions that are 100% shared by the whole set of 287 pigs. The shared functions
are consistent with those identified for the entire pig gut gene catalogue both in
relation to annotation and relative abundances.
Distribution of KOs shared by all the pig samples
Function c
lass
Organismal Systems −− Sensory SystemOrganismal Systems −− Nervous SystemOrganismal Systems −− Immune System
Organismal Systems −− Excretory SystemOrganismal Systems −− Environmental Adaptation
Organismal Systems −− Endocrine SystemOrganismal Systems −− Digestive System
Organismal Systems −− DevelopmentOrganismal Systems −− Circulatory System
Metabolism −− Xenobiotics Biodegradation and MetabolismMetabolism −− Nucleotide Metabolism
Metabolism −− Metabolism of Terpenoids and PolyketidesMetabolism −− Metabolism of Other Amino Acids
Metabolism −− Metabolism of Cofactors and VitaminsMetabolism −− Lipid Metabolism
Metabolism −− Glycan Biosynthesis and MetabolismMetabolism −− Enzyme Families
Metabolism −− Energy MetabolismMetabolism −− Carbohydrate Metabolism
Metabolism −− Biosynthesis of Other Secondary MetabolitesMetabolism −− Amino Acid Metabolism
Human Diseases −− Neurodegenerative DiseasesHuman Diseases −− Metabolic DiseasesHuman Diseases −− Infectious Diseases
Human Diseases −− Immune System DiseasesHuman Diseases −− Cardiovascular Diseases
Human Diseases −− CancersGenetic Information Processing −− Translation
Genetic Information Processing −− TranscriptionGenetic Information Processing −− Replication and Repair
Genetic Information Processing −− Folding, Sorting and Degradation Environmental Information Processing −− Signaling Molecules and Interaction Environmental Information Processing −− Signal Transduction Environmental
Environmental Information Processing −− Membrane TransportCellular Processes −− Transport and Catabolism
Cellular Processes −− Cell MotilityCellular Processes −− Cell Growth and Death
Cellular Processes −− Cell CommunicationF
unction c
lass
Distribution of KOs from annotated species
0e+001e+055e+05 4e+05 3e+05 Number of matched genes
2e+056e+057e+05 0e+00 1e+05 2e+05 3e+05 4e+05 Number of matched genes
5e+05 6e+05 7e+05
Supplementary Figure 3
Supplementary Figure 4 | Functional comparison of the pig, human and mouse
catalogues. The Venn diagram (a) provides the number of shared and species-specific
KEGG pathways for the pig (pink), the mouse (grey), and the human (yellow)
catalogues. The classification of the 2179 KEGG pathways found 100% shared by the
three animal species (b) highlights the predominance of common metabolic functions
related to carbohydrates and amino acids as well as environmental information
processing (membrane transport), consistent with the most abundant functions found
in the pig catalogue (see Supplementary Fig. 2 and Supplementary Fig. 3).
a
b
Supplementary Figure 4
Supplementary Figure 5 | Effects of host genetics on the pig microbiome
composition. The influence of host genetics was assessed by NMDS from the subset
of Chinese pigs, at the levels of phylum (a), genus (b), species (c), MGS (d) and
KEGG pathways (e). The MGS-based NMDS clearly distinguished three groups
corresponding to the highly selected commercial breed (HybCN1, HybCN2 and Large
White), the Bama and related BaRing pigs, and the Tibetan pigs. The KEGG-based
NMDS still clearly separated the Tibetan pigs from the others, suggesting specific
microbiota functions in this breed.
a b c
d e
Supplementary Figure 5
Supplementary Figure 6 | Age effect on the pig gut microbiota composition. The
influence of age was assessed by NMDS at the levels of the total NR gene counts (a)
and KEGG pathways (b, c), from the subsets of Danish pigs (a, c) and French pigs (b)
(counterpart of Fig. 3b in the main text). The animals are distributed in a surface plot
that includes lines referring to their ages, showing an effect of age, likely to be
connected to the diet and environmental changes during lifetime.
Supplementary Figure 6
a b c
Supplementary Figure 7 | Effect of the farm system (country) on the gut
microbiota composition. NMDS based on the NR gene counts (a), the KEGG
pathways (b) and the MGS counts revealed a separation between the Chinese pigs
CP), and the French (FP) and Danish (DP) pigs. This separation is likely to be partly
related to different farm systems. In addition, the French pigs comprised 11 subsets
divided by the breed, the environment, or both, resulting in more diversity than for the
Danish pigs, as revealed by this analysis.
Supplementary F igure 7
c
a b
Supplementary Figure 8 | Country-specific relative abundances of subsets of
antibiotic resistance genes (ARGs) and of KOs related to the tricarboxylic acid
(TCA) cycle. (a) Aminoglycosides; (b) Beta-lactams; (c) Sulfonamide; (d) KOs in
the TCA cycle.
Beta-lactams
Aminoglycosides
Sulfonamide
A
B
C
D a Aminoglycosides
b Beta-lactams
c Sulfonamide
d
Supplementary Figure 8
Supplementary table 1. Background information on the 287 pig samples.
Supplementary table 2. Description of the assembly data from the 287 samples.
Supplementary table 3. Assembly results of the pig, human and mouse data.
Supplementary table 4. Significant differences in abundance between the gut
microbiomes of castrated males and females (10 castrated males, 10 females,
Svindinge farm) at the genus and species levels.
Supplementary table 5. Significant differences in abundance between the gut
microbiomes of castrated males and females (10 castrated males, 10 females,
Svindinge farm) at the KEGG pathway level.
Supplementary table 6. Significant differences in abundance between the gut
microbiomes of males and females (11 males, 14 females, fed wet feed, Stærsminde
farm) at the genus and species levels.
Supplementary table 7. Significant differences in the abundance of MGS in the gut
microbiomes of males and females (11 males, 14 females, fed wet feed, Stærsminde
farm).
Supplementary table 8. Significant differences in KEGG pathways in the gut
microbiota of males and females (11 males, 14 females, fed wet feed, Stærsminde
farm).
Supplementary table 9. List of the KEGG pathways mapped by iPATH2 that differed significantly between the gut microbiota of males and females (11 males, 14 females, fed wet feed, Stærsminde farm). The three most represented pathways are highlighted in grey. This list is derived from table S8 and corresponds to the differentially abundant functions mapping against KEGG according to iPATH2 tools.