11/16/2012
Microbial communities often exhibit predictable patterns in
composition and diversity across environmental gradients
The promise (and peril) of using comparative
shotgun metagenomics to understand
belowground communities
Noah Fierer
Assoc. Professor
Dept. of Ecology & Evolutionary Biology
Univ. of Colorado at Boulder
Soil pH
Soil pH
Lauber et al. 2009. Appl. Environ. Micro.
Alpha Diversity
Mantel r (soil pH) = 0.81
Ladau et al. In Prep.
Changes in microbial
community structure
meta-‘omics’
Griffiths et al. 2010. Environ. Micro.
Changes in soil
processes
Can we use soil metagenomics to improve our
understanding of microbial community function?
map of 16 sites
with legend,
include photos
/
1
11/16/2012
abundant cyanobacteria
Shotgun metagenomic sequencing
2400
4-11 million reads per sample (Illumina HiSeq)
Tropical Forest
Temperate Coniferous Forest
Temperate Deciduous Forest
Boreal Forest
Prairie
Arctic Tundra
Hot Desert
Cold Desert
Metagenomic richness
2200
≈ 20% of reads could be annotated
data rarefied to 688,000 annotated reads/sample
2000
1800
1600
1400
≈ 85% of LSU or SSU rRNA genes assigned to bacteria
r2 = 0.49
1200
200
300
400
500
600
700
16S rRNA phylogenetic diversity (PD)
16S rRNA gene amplicon sequencing
118,000 reads per sample (Illumina HiSeq)
Fierer et al. In Review.
40
16S rRNA genes
20
Tropical Forest
PCo2 (15%)
Temperate Coniferous Forest
Temperate Deciduous Forest
0
Boreal Forest
Prairie
Arctic Tundra
-20
Hot Desert
Cold Desert
-40
β diversity – 16S rRNA amplicon results
-60
-60
β diversity – shotgun metagenomes
-40
-20
0
20
40
60
PCo1 (26%)
0.4
1.5
metagenomes
1.0
0.2
Tropical Forest
Tropical Forest
0.5
Temperate Deciduous Forest
Boreal Forest
PCo2 (11%)
PCo2 (14%)
Temperate Coniferous Forest
0.0
Prairie
Arctic Tundra
Hot Desert
-0.2
Cold Desert
Temperate Coniferous Forest
Temperate Deciduous Forest
0.0
Boreal Forest
Prairie
-0.5
Arctic Tundra
Hot Desert
Cold Desert
-1.0
-0.4
-1.5
-0.4
-0.3
-0.2
-0.1
0.0
0.1
0.2
0.3
0.4
-2.0
PCo1 (25%)
-4
-3
-2
-1
0
1
2
3
4
PCo1 (71%)
desert
40
other biomes
desert
Fierer et al. In Review.
rRNA genes
other16Sbiomes
Fierer et al. In Review.
20
40
Tropical Forest
16S rRNA genes
20
Tropical Forest
PCo2 (15%)
Temperate Coniferous Forest
Temperate Deciduous Forest
0
PCo2 (15%)
Temperate Coniferous Forest
Temperate Deciduous Forest
0
Boreal Forest
Prairie
Arctic Tundra
-20
Hot Desert
Boreal Forest
Cold Desert
Prairie
Arctic Tundra
-20
-40
Hot Desert
Cold Desert
-40
-60
-60
-40
-20
0
Community Composition (16S)
0.4
Functional Attributes
(shotgun)
PCo1 (26%)
-60
-40
-20
0
20
40
1.5
metagenomes
1.0
Tropical Forest
1.0
Tropical Forest
0
-20
-0.2
Tropical Forest
Temperate
Coniferous Forest
0.5
Tropical Forest
Temperate Coniferous Forest
Temperate Deciduous Forest
Temperate Coniferous Forest
Temperate Deciduous Forest
PCo2 (11%)
PCo2 (15%)
PCo2 (14%)
0.0
0.5
Boreal
0.0Forest
Boreal Forest
Prairie
Prairie
Arctic Tundra
Arctic Tundra
Hot Desert
Hot Desert
Cold Desert
Cold Desert
-0.5
Temperate Deciduous Forest
Boreal Forest
Prairie
Arctic Tundra
Hot Desert
Temperate Coniferous Forest
Temperate Deciduous Forest
0.0
Boreal Forest
Prairie
-0.5
Arctic Tundra
Hot Desert
Cold Desert
-1.0
Cold Desert
-1.0
-40
-1.5
-0.4
-0.4
PCo2 (11%)
20
60
metagenomes
16S rRNA genes
0.2
40
60
1.5
40
20
PCo1 (26%)
-60
-1.5
-0.3
-0.2
-0.1
-2.0
-60
0.1 -60 0.2
0.0
-400.3
-20
0.4
0
PCo1 (25%)
20
-2.0
PCo1 (26%)
-4
40
-3
-4
60
-2
-1
0
1
2
3
-3
-2
-1
4
0
1
2
3
4
PCo1 (71%)
PCo1 (71%)
1.5
metagenomes
1.0
Tropical Forest
0.5
PCo2 (11%)
Community composition is strongly
correlated with functional attributes
(Mantel r = 0.76)
Temperate Coniferous Forest
Temperate Deciduous Forest
0.0
Boreal Forest
Prairie
-0.5
Arctic Tundra
Hot Desert
Cold Desert
-1.0
-1.5
Deserts
 dormancy/sporulation
 osmoregulation
 amino acid metabolism
Other Biomes
 catabolism of aromatic C
 N, P, S cycling
 antibiotic resistance
-2.0
-4
-3
-2
-1
0
1
2
3
4
PCo1 (71%)
2
11/16/2012
Antibiotic Resistance Genes
Other Biomes
Relative Abundance of Antibiotic
Resistance Genes (%)
6
Strong desert/non-desert split in the functional
attributes of soil microbial communities
5
Soil microbial biogeography is often predictable
4
3
Cold
Deserts
Hot
Deserts
- diversity
- community composition
- functional traits
2
1
0
1
6
3
4
1
0
9
7
6
2
3
1
1
1
1
1
02 02 02 01 01 02 MD SV SF PE KP DF CL BZ AR TL
E B E B E B EB E B E B
Fierer et al. In Review.
Effects of nitrogen fertilization on soil carbon dynamics
+N
Effect Size
Meta-analysis of previously published field studies
plant biomass
lower
higher
microbial biomass
higher
lower
microbial respiration rates
Soil microbial biomass and respiration typically decrease
when soils are amended with N
Ramirez et al. 2012. Global Change Biol., Treseder. 2008. Ecology Letters
3
A
0
100 200 300 400 800
500
Control Respiration (mg C g C-1)
Paired samples: control and treatment (1mg N/g soil)
0.0
Effect Size
30% decrease in
microbial biomass
Lab Incubation (365 days)
0
20
40
60
80 100
Control Biomass (mg C g C-1 h-1)
0.1
10% decrease in
soil respiration
B
-0.4 -0.3 -0.2 -0.1
Cross Biome Study:
Treatment Biomass (mg C g C-1 h-1)
0 20 40 60 80
Treatment Respiration (mg C g C-1)
0 100
300
500
800
11/16/2012
Respiration
Ramirez et al. 2012. Global Change Biol.
Biomass
Ramirez et al. 2012. Global Change Biol.
Why does microbial biomass and respiration typically
decrease when soils are amended with N?
Fertilizer toxicity? (unlikely - see Ramirez et al. 2010. Soil Biol. Biochem.)
Experimental N gradients
Cedar Creek LTER, MN
Kellogg Biological Station LTER, MI
N additions in place for 28 years
(up to 272 kg/ha/yr)
N additions in place for 10 years
(up to 290 kg/ha/yr)
Inhibition of extracellular enzymes?
Change in microbial community composition?
16S rRNA analyses (Unifrac distances)
N increases
‘copiotrophic’ taxa
Bacteroidetes
γ-Proteobacteria
Shotgun metagenomic analyses
18 samples: 2 sites x 3 N levels x 3 reps.
Cedar Creek
Kellogg
high N
1.3 million reads (454 Titanium)
15,000 to 28,000 annotated reads per sample
N decreases
‘oligotrophic’ taxa
Acidobacteria
low N
Ramirez et al. 2011. Ecology
Fierer et al. 2012. ISME J
4
11/16/2012
The composition of the communities and their
functional attributes are well-correlated
CC r2 = 0.98
KBS r2 = 0.60
CC no N
CC high N
KBS no N
KBS med N
KBS high N
CC no N
CC med N
MG-RAST PCo1
CC med N
CC high N
KBS no N
KBS med N
KBS high N
Unifrac PCo1
Fierer et al. 2012. ISME J
A
-1
0 kg N ha y
Subsystem Hierarchy 1
Subsystem Name
Subsystem
#
Fierer et al.
2012. ISME J
-1
34 kg N ha y
-1
-1
272 kg N ha y
-1
45
54
8
23
38
17
40
52
Rho
Mean
%
-0.4
-0.4
-0.8
-0.5
-1.3
1.0
1.6
1.2
0.91
1.2
-0.1
-0.4
-1.2
-0.4
-0.8
-0.8
1.3
0.8
1.6
0.91
0.3
-0.9
-0.5
-1.1
-0.7
-0.4
-0.2
1.2
1.6
1.1
0.97
0.1
139
-0.9
-0.7
0.1
-0.6
-0.9
-0.7
0.7
1.4
1.6
0.90
4.3
143
-0.9
-1.2
-0.3
-0.1
0.4
-1.0
1.9
0.4
0.8
0.81
0.5
-0.3
-0.8
-1.1
-0.4
0.1
-1.0
0.6
1.4
1.6
0.90
0.4
155
F0F1-type ATP synthase
154
Cofactors, Vitamins
Lipoic acid metabolism
96
Protein Metabolism
tRNA aminoacylation
Protein Metabolism
General Secretion Pathway
DNA Metabolism
DNA structural proteins
106
Respiration
-1
22
-0.3
Ubiquinone-cytochrome reductase
complexes
Respiration
DNA Metabolism
DNA-replication
104
-0.9
-1.4
-0.7
-0.8
0.0
0.6
1.5
1.2
0.3
0.80
1.4
RNA Metabolism
ATP-dependent RNA helicases
145
-0.7
-0.1
-0.1
-0.3
-1.0
-0.4
1.9
1.5
-0.8
0.83
1.5
Nucleosides/Nucleotides
Purine conversions
128
-1.4
-0.9
-0.8
0.3
0.3
-0.6
0.6
0.7
1.8
0.62
1.1
Amino Acids and Derivatives
Urea decomposition
2
2.0
0.5
-0.1
-0.8
0.5
0.6
-0.8
-0.9
-1.0
-0.70
1.2
Motility and Chemotaxis
Bacterial Chemotaxis
125
1.1
-0.4
0.6
1.4
0.5
-0.6
-1.7
-0.1
-0.9
-0.67
1.3
Motility and Chemotaxis
Bacterial motility:Gliding
124
-0.1
0.3
1.1
1.7
-0.7
0.6
-0.7
-0.8
-1.3
-0.70
0.6
Aromatic Cmpd. Metabolism
Phenylpropanoid compound
degradation
120
0.2
2.0
-0.4
-0.2
0.7
0.7
-1.1
-1.1
-0.6
-0.73
0.6
Cofactors, Vitamins
Fe-S cluster assembly
93
1.6
1.1
0.0
0.7
0.0
-0.1
-0.8
-1.1
-1.3
-0.84
Regulation and Cell Signaling
cAMP signaling
153
1.1
1.1
1.1
-0.1
0.2
0.1
-1.4
-1.4
-0.6
-0.91
Clustering-based Subsystems
Tricarboxylate transporter
80
0.4
1.0
0.6
0.5
0.4
0.8
-1.9
-1.1
-0.8
-0.94
0.5
Rho
Mean
%
B
-1
0 kg N ha y
Subsystem Hierarchy 1
Subsystem Name
Respiration
Ubiquinone-cytochrome reductase
complexes
Cofactors, Vitamins
Folate Biosynthesis
-1
-1
101 kg N ha y
-1
-1
291 kg N ha y
201
301 401 204 304
404
209
309
409
155
-1.3
-0.1
-0.5
-0.2
1.3
2.0
94
-0.8
0.0
-1.5
0.9
1.2
Subsystem#
-0.4
-0.6
-0.2
0.1
0.76
-1.1
-0.1
1.3
0.86
Lipoprotein Biosynthesis
138
-1.7
-0.3
0.2
-0.2
-1.0
-0.1
1.2
1.6
0.3
0.74
RNA Metabolism
ATP-dependent RNA helicases
145
-1.0
-0.6
0.2
-1.2
-0.5
0.6
-0.1
2.0
0.7
0.62
Cell Wall and Capsule
Cellulosome
31
-0.7
-1.1
-0.4
-0.2
-0.3
-0.5
-0.1
1.7
1.7
0.82
Potassium metabolism
Potassium-efflux system
135
-0.9
-0.1
-1.0
-0.8
0.4
-0.5
0.9
0.0
2.1
0.74
Protein Metabolism
0.1
2.5
-1
Clustering-based subsystems
Pyrimidine biosynthesis
62
-1.7
0.3
-1.5
-0.2
0.2
0.5
0.8
0.2
1.4
0.74
Membrane Transport
Fructose and Mannose Inducible PTS
115
-0.9
-0.7
-0.2
0.2
0.1
-1.5
0.0
1.2
1.7
0.71
Carbohydrates
Chitin and N-acetylglucosamine
utilization
11
-0.2
-0.2
-0.4
-1.0
-1.1
-0.3
-0.1
1.7
1.7
Amino Acids and Derivatives
Urea decomposition
2
1.9
-0.7
0.7
0.2
-0.3
0.7
-1.3
0.1
-1.1
Secondary Metabolism
Phytoalexin biosynthesis
163
1.5
1.0
1.0
-0.7
-0.4
-0.6
-1.5
0.2
-0.5
-0.69
Nucleosides and Nucleotides
Pyrimidine utilization
129
1.0
0.4
0.7
-0.2
0.3
0.7
-2.2
0.3
-0.9
-0.73
Clustering-based subsystems
Tricarboxylate transporter
80
-0.4
1.1
0.3
0.9
0.5
1.1
-1.2
-1.4
-0.9
-0.74
Cell Wall and Capsule
Lipid A-Ara4N pathway (Gram
negative)
29
0.9
1.6
0.2
-0.1
0.7
-0.4
-0.6
-1.7
-0.7
Cell Wall and Capsule
Rhamnose containing glycans
27
1.4
1.0
1.2
-0.6
0.4
-0.6
-0.4
-1.0
-1.3
-0.85
Virulence
Resistance to fluoroquinolones
189
1.9
0.4
0.8
0.2
-0.4
0.2
-0.8
-0.9
-1.4
-0.88
0.67
-0.61
-0.83
1.2
1.5
0.1
1.6
0.1
0.9
0.1
0.1
0.2
1.3
0.1
0.4
0.7
1.2
1.2
3.2
Shotgun metagenomic analyses indicate that N increases
“copiotrophic” gene pathways
20
Caveats
Metagenomics provides information on potential functions
No added N
+N
Parting Thoughts:
Microbial community composition does provide
information on ecological attributes
>75% of the soil metagenome is typically left unannotated
Gene sequence homology may not equal functional homology
We need more genomic and ecological information for the
majority of soil microbial taxa that remain uncharacterized
Gene relative abundance ≠ absolute abundance (or activity)
Meta-‘omics’ are no panacea
Fungi likely underrepresented in metagenomes
5
11/16/2012
Project Collaborators
Univ. of Colorado:
Scott Bates
Donna Berg-Lyons
Rob Knight
Chris Lauber
Jon Leff
Jesse Zaneveld
Colorado State Univ.:
Kelly Ramirez
Diana Wall
Northern Arizona Univ.:
Greg Caporaso
Yale Univ.:
Mark Bradford
Univ. of California, San Francisco:
Josh Ladau
Katie Pollard
Brigham Young Univ.:
Byron Adams
Univ. of Western Sydney:
Uffe Nielsen
Argonne National Lab/Univ. of Chicago:
Jack Gilbert
Sarah Owens
U.S. Dept. of Agriculture
National Science Foundation
6