Mock Bp LPS Bp338 BpMinn1 E. coli S. aureus-1 S. aureus-2 ...genome- · Mock Bp LPS Bp338 BpMinn1 E. coli S. aureus-1 S. aureus-2 Iono+PMA BTG2 Similar to sgk sgk c-jun c-fos c-fos

Mock Bp LPSBp338 BpMinn1

E. coliS. aureus-1

S. aureus-2Iono+PMA

BTG2 Similar to sgk

sgk

c-jun c-fos c-fos

Unknown Hs.123581 ESTs

BTG2 EGR-1=Early growth response protein 1 AREB6=zinc finger homeodomain protein PrP=prion protein Hs.158876 EST Similar to MxB SLAP=src-like adapter protein SLAP=src-like adapter protein

LEC=liver CC chemokine-1 PDGF2=platelet-derived growth factor 2

Hs.69298 ESTs Hs.69298 ESTs

Signal recognition particle 72 (SRP72) Thioredoxin reductase Topoisomerase I GADD45 alpha KIAA0838=Similar to glutaminase Cyclin D2 MMP-1

Serine hydroxymethyl transferase Gfi-1 CTP synthetase

Similar to HEM45 cdk4 Glycyl tRNA synthetase GRSF-1=cytoplasmic G-rich mRNA sequence binding factor dystrobrevin B DTN-B2=dystrophin-associated protein A0 ornithine decarboxylase 1=odc1 CTP synthetase RAN=GTP-binding nuclear protein G1 to S phase transition 1 Protein synthesis factor (eIF-4C) MAPKAP kinase (3pK) Eukaryotic translation initiation factor 4E

phosphoglycerate mutase brain isoform (PGAM1)

Gfi-1=growth factor independence-1=zinc finger protein methionine adenosyltransferase alpha subunit uridine phosphorylase HSP70-HOM=Heat shock 70 KD protein 1 HSP70

STP1=phenol-preferring phenol sulfotransferase1 P-selectin glycoprotein ligand

gamma-interferon inducible gene IP-30

HMG-2=high mobility group 2 protein Net=ELK3

R-PTP-alpha=receptor protein tyrosine phosphatase alpha CD49C=Integrin alpha 3 Arachidonate 5-lipoxygenase=5-lipoxygenase=5-LO MxB c-myc

SIP-110=signaling inositol polyphosphate 5 phosphatase WIP/HS PRPL-2=WASP interacting protein CD49C=Integrin alpha 3 PKC beta =Protein kinase C, beta Tis11d=ERF-2=growth factor early response gene ELF-1=ets family transcription factor SH3P18=SH3 domain-containing protein PCM-1=autoantigen pericentriol material 1 pLK=homologue of Drosophila polo serine/threonine kinase CD105=endoglin Cytotoxic ligand TRAIL receptor

CD103 alpha=Integrin alpha-E ATM GRK5=G protein-coupled receptor kinase 5

Glutathione peroxidase 1

IL-16 CD62L=L-selectin=LAM-1

c-myb CD31=PECAM-1 CD33 SLAP=src-like adapter protein CD11A=Integrin, alpha L=LFA-1 alpha chain Transforming growth factor, beta receptor

PARP MDR1=Multidrug resistance-associated protein 1

calcium/calmodulin-dependent protein kinase II

CD64 TTG-2 C-C chemokine receptor 1=CC CK1 HNPP=nuclear phosphoprotein

MAPKAP kinase (3pK) cytohesin-1 CD115=CSF-1 receptor thymosin beta-4 Myeloid cell nuclear differentiation antigen

Peroxisome assembly factor-2 (PAF-2)

Peroxisome biogenesis disorder protein 1 (PEX1)

UntreatedBp338

Bp537BpA2-6

BpTox6

>4X

2X

Unchanged

2X

>4X

FoldInduced

FoldRepressed

GO# Annotation Term Example Induced Genes p value 6960 Anti-apoptosis c-IAP2, Bfl-1, Bcl-XL 8.3 x 10-3 7276 Antimicrobial humoral response SLAM, CD40, Adenosine deaminase 9.0 x 10-6

2125 Cell-cell signaling TNFα, IL6, MIP1α, MIP1β 2.9 x 10-6

5615 Chemokine PARC, LARC, LD78β 8.3 x 10-3

7165 Cytokine IL1α, IL1β, IL10 2.6 x 10-4

6916 Extracellular space TGFα, G-CSF, HDGF 1.0 x 10-3

6960 Signal transduction STAT3, CD69, Thioredoxin, CREB 5.2 x 10-4

GO# Annotation Term Example Repressed Genes p value 6954 Inflammatory response 5-LO, CD14, CYBB, CCR1, FPR 1.0 x 10-5 9596 Perception of pest/pathogen/parasite MHC II transactivator, HLA-DMα, HLA-DMβ 1.0 x 10-2

5886 Plasma membrane TNFR1, CD13, CD31, RGS14, CD105 2.9 x 10-4

Table 1. Enrichment of Gene Ontology (GO) annotation terms in the bacterial diversity data set

( ) ∑−

=

−−

−=1

0

1 ,,;x

i

nN

inAN

iA

nANxp , where ( )! !/! nNnNnN −=

.

Using the bacterial diversity data set, 1,208 unique genes were selected based on having a Locus

Link assignment (http://www.ncbi.nlm.nih.gov/LocusLink). p values for the frequency of 184 GO

annotation terms (those that had at least 5 occurrences in the data set) were calculated based on the

hypergeometric distribution. Those annotation terms found to have p values less than 1% within the

common induction (red) and common repression (green) clusters are displayed with their GO identification

numbers and example genes from the 1,208-gene data set. The GO annotations for genes in the entire

bacterial diversity data set were extracted in batch from SOURCE

(http://genome-www.stanford.edu/source), using the Locus Link identification number for each gene. Th e

resulting 1,208 unique annotated genes were used for the analysis, among which 184 annotation terms

occurred for at least 5 genes. We compared the frequency of these 184 annotations in the data set as a

whole to that in the common induction and common repression clusters, using the hypergeometric

distribution to calculate p values (Tavazoie, S., Hughes, J. D., Campbell, M. J., Cho, R. J. & Church, G. M.

(1999) Nat. Genet. 22, 281-285) (see equation below). Let N = 1208 denote the total number of genes

under consideration and A, the number of these genes with a particular annotation. The chance of

observing at least x genes with that annotation in a random subset of n genes is given by

Detection of bacteria/bacterial products

Systemic effects

Cell-to-cell signalingIntracellular signalling

Fever (IL-1, IL-6)Mobilization of metabolites (TNFα)Shock (TNFα)Production of acute-phase proteins (IL-6)

Activation (IL-1, IL-6, TNFα)Stimulation of differentiation (GM-CSF, LIF)Chemoattraction (MIP-1, MIP-2, PARC)

+/- regulation of NF-kB signalling pathway (NFkB1, Rel, IkB)Metabolic processes (Adenosine deaminase, Thioredoxin)+/- regulation of apoptosis (cIAP, LTA)

Common inductiona

Recognition & destruction of microbeCD64, WASP, CYBB,

NCF-1

Chemotaxis to infection site

Presentation of microbial peptidesCathepsin B, MHC Class II, IP30,

CD13

MCP-1, n-Formyl peptide receptor,CCR1, CD31

Common repressionb

0 1 2.3

lowabsolute

difference

highabsolute

difference

r=0.85

IL6

Tim

e (h

ours

)

2.04.06.08.01012

0.5

Dose10X 100X 1000X1X

Tim

e (h

ours

)

2.04.06.08.01012

0.5

Dose10X 100X 1000X1X

IL1β

r=0.86

r=0.75

Defensin, alpha 1

Dose10X 100X 1000X1X

Tim

e (h

ours

)

2.04.06.08.01012

0.5

Tim

e (h

ours

)

2.04.06.08.01012

0.5

Dose10X 100X 1000X1X

CXCR4

r=0.77

IL1α

r=0.76Dose

10X 100X 1000X1X

Tim

e (h

ours

)

2.04.06.08.01012

0.5

r=0.79

Mip2α

Dose10X 100X 1000X1X

Tim

e (h

ours

)

2.04.06.08.01012

0.5

r=0.87

CD31

Dose10X 100X 1000X1X

Tim

e (h

ours

)

2.04.06.08.01012

0.5

r=0.87

SLAM

Dose10X 100X 1000X1X

Tim

e (h

ours

)

2.04.06.08.01012

0.5

Unknownchitinase precursorAPR=immediate-early-response geneRGS1=regulator of G-protein signalingCD69=early activation antigenAPR=immediate-early-response geneUnknownSimilar to CD95=FasIL-15 receptor alpha chainThioredoxinHs.172051 ESTsHs.105499 ESTsHs.125887 ESTsSimilar to polyprotein-encoding gene 1IRF-7=interferon regulatory factor-7Hs.104444 ESTsHs.230206 ESTsPMLStaf50HEM45=interferon-inducible PML nuclear bodies-associated proteinGlucocorticoid receptorHs.123664 ESTsIFI16=interferon-gamma-inducible myeloid differentiation transcriptional activatorIFI16=interferon-gamma-inducible myeloid differentiation transcriptional activatorMxB=interferon-induced cellular resistance mediator proteinIFP35=interferon-induced leucine zipper proteinUnknownUnknown sc_id = 49512’-5’ oligoadenylate synthetase E2’-5’ oligoadenylate synthetase EHs.56009 ESTsHs.56009 ESTsHs.82554 ESTsHEM45=gpISG20=interferon-inducible PML nuclear bodies-associated proteinHEM45=gpISG20=interferon-inducible PML nuclear bodies-associated proteinTRAIL=Apo-2 ligandSimilar to 5’-3’ exonuclease=cytoplasmic exoribonuclease mXRN1pHomo sapiens mRNA for deoxyribonuclease III drn3 geneIP-10Similar to myosin-binding protein C slow-type muscleMIF=macrophage migration inhibitory factorAIM2=interferon-inducible protein

Mock Bp LPSBp338

BpMinn1E. co

li(1)S. aureus(1)

S. aureus(2)

Iono+PMABp338 Dose S. aureus Dose Bp LPS Dose

0.01ug/mlMock

0.1ug/ml

1.0ug/ml1X 10X 100X

1000X 1X 10X 100X1000X

KilledLiveBp338

E. coli(2)

Dose Response Dataset Diversity DatasetLive vs. KilledDataset

Supplemental Tables & Figures: (Enhanced versions also available at http://relman.stanford.edu/hostresponse) Supplemental Figure i: A subset of transcripts are induced exclusively by ionomycin/PMA treatment. Data are the same as in Fig. 1a. Transcripts with multiple copies indicate multiple measurements. Supplemental Figure ii: A subset of genes in the common repression response are similarly repressed during B. pertussis infection of U937 cells. Macrophages differentiated from the U937 cell line were infected with B. pertussis virulent (Bp338), avirulent (Bp537), and isogenic mutant strains (BpA2-6, BpTox6) as in Fig. 4c. Data were selected based on the same criteria as the bacterial diversity and dose response data sets, except genes were filtered based on expression of at least 1.5-fold below the mean for at least four of the sampled time points. Genes that overlapped in expression with the common repression cluster from the bacterial diversity data set are highlighted in red. Supplemental Table iii: Enrichment of GO annotation terms in the bacterial diversity data set. Using the bacterial diversity data set, 1208 unique genes were selected based on having a LocusLink assignment ((http://www.ncbi.nlm.nih.gov/LocusLink). p-values for the frequency of 184 GO annotation terms (those that had at least 5 occurrences in the data set) were calculated based on the hypergeometric distribution. Those annotation terms found to have p-values less than 1% within the Common Induction (red) and Common Repression (green) clusters are displayed with their GO identification numbers and example genes from the 1208-gene data set. The Gene Ontology annotations for genes in the entire bacterial diversity data set were extracted in batch from SOURCE (http://genome-www.stanford.edu/source), using the Locus Link identification number for each gene. The resulting 1208 unique annotated genes were used for the analysis, among which 184 annotation terms occurred for at least 5 genes. We compared the frequency of these 184 annotations in the data set as a whole to that in the common induction and common repression clusters, using the hypergeometric distribution to calculate p-values (Tavazoie, 1999 Nature Genetics) (see equation below). Let N=1208 denote the total number of genes under consideration and A the number of these genes with a particular annotation. The chance of observing at least x genes with that annotation in a random subset

of n genes is given by ( ) ∑−

=

−−

−=1

01,,;

x

i

nN

inAN

iA

nANxp , where ( )!!/! nNnNnN

−=

.

Supplemental Figure iv: Physiological programs represented in the common response. Physiological programs represented in the common induction (a) and common repression (b) response clusters. Representative examples in the “common induction” or “common repression” clusters associated with specific features of the physiological program of innate immunity are indicated. Supplemental Figure v: Modeling of the dose response reveals discrimination. The response difference profiles (as in the fourth panel of Fig. 3) for 8 different genes are displayed in 2-dimensional format. The correlation of the fitted response to the observed response is displayed below each of the plots. The x-axis represents time, the y-axis, exposure dose, and the z-axis, the expression level (log2) of the zero-transformed data. The surface represents the least squares fit of the expression data to the model. The subtraction of the two expression surfaces as defined by delta(d,t). Supplemental Figure vi: A subset of genes varies in expression among the donors. As described in the text (Donor expression differences), the expression profiles of a cluster of genes identified in the bacterial diversity dataset (Fig.1a, gray bar), was assessed in the response of 3 separate donors, including a replicate experiment to verify the differential response of the first donor to E. coli.