ulementary inormation - media.nature.com · 215 m l. a . ulementary inormation. 2. ... tuba4a ppia...

S U P P L E M E N TA RY I N F O R M AT I O N

WWW.NATURE.COM/NATURECELLBIOLOGY 1

DOI: 10.1038/ncb3257

a b

d

Supplementary Figure 1 (Humphries)

−1

1

−1 0 1

PC2

K562

MEF

HFF

A375

MKF1

MKF2

MKF3 PC1

K56

2

MEF

A37

5

HFF

MKF

1

MKF

2

MKF

3

Adh

esom

e

K562 MEF A375 HFF MKF1 MKF2 MKF3 Adhesome

402674

1,023314

392754

654232

115

≥0.20

Jaccardcoefficient

166 53 56 62 60 21270 100 116 145 123 53

98 98 145 102 75101 143 140 46

266 247 40413 58

47

c

≥2<2

Enrichment(FN/control)

Prot

eins

Datasets

HFF

MKF

1

MKF

2

MKF

3

K562

MEF

A375

0.5

0.6

0.7

0.8

0.9

Jacc

ard

dist

ance

1.0

Occ

urre

nce

Adhe

som

e

1 7

Datasetoccurrence

Supplementary Figure 1 Comparison of FN-enriched IAC proteomes. (a) Seven proteomic datasets of FN-enriched IACs were analysed by unsupervised hierarchical clustering. The binary heat map shows proteins at least two-fold enriched to FN over the negative control (red). Dataset occurrence is plotted for each protein (rainbow), and literature-curated adhesome4 components are indicated by purple bars. Details of the proteomic datasets are provided in Supplementary Table 1. (b) Dendogram illustrating the clustering of the FN-enriched IAC proteomes shown in a. Dataset dissimilarity is measured by Jaccard distance. (c) Pairwise overlaps of FN-enriched proteins identified in the seven proteomic datasets and

the literature-curated adhesome were measured by Jaccard coefficient and are displayed as a hierarchically clustered heatmap (lower diagonal matrix; blue). Numbers of proteins in each overlap set are indicated (upper diagonal matrix). (d) FN-enriched proteins identified in the seven proteomic datasets were analysed by principal component analysis. A plot of the first two principal components is shown. K562, human chronic myelogenous leukaemia cells11; MEF, mouse embryonic fibroblast cells (this study); A375, human malignant melanoma cells14; HFF, human foreskin fibroblast cells13; MKF1, mouse kidney fibroblast cells15; MKF2 and MKF3, mouse kidney fibroblast cells16.

© 2015 Macmillan Publishers Limited. All rights reserved

http://www.nature.com/doifinder/10.1038/ncb3257


2 WWW.NATURE.COM/NATURECELLBIOLOGY

a


ARHGEF6

PRKCDRDX

DNM2

FLNA

SORBS3

SORBS1PTPN11SORBS2

ITGB4

MYH9

ASAP2SVIL

GIT1

FHL2

CBLSRC

LYN

RHOACRK

RAC1SH3KBP1 PLEC

ACTN1GIT2

MARCKS

ACTB

CFL1

PRNP

PTPN2

TRIP6VCL

MSN

STAT3

KTN1

NUDT16L1VAV2

PARVA

LIMS2

MACF1HSPA2HSPB1

ILK

TES

PAK1

TUBA1B

VASP

OSTF1

RAPGEF1ENAHARHGAP24

INPPL1

PLCG1

SH2B1

FBLIM1FERMT2

GRB2

CAPN2

ABI1

CTTN

NCK2

PTPN1

PTPRA

BCAR1

PPP2CA

CAV1

ROCK1

ARHGEF7

PPFIA1

IRS1ARHGEF2

PPM1F

EZR

ITGB3

PALLD

ITGA6

TENC1

ARHGAP35

MAPK1

PTPN12

TGFB1I1ITGA4

ITGA2ITGAV

CALR

LIMS1

ITGA3

PTK2

ZYX

LPXN

ITGB5

ITGB1

CRKL

PXN

PIK3CACSK

RASA1CSRP1

TLN1

LASP1

ITGA5

TUBA4ASLC3A2

ARPC2

VIM

LPPPARVB

1 2 3 4 5 6 710−710−610−510−410−310−210−1100

Dataset occurrence

Betw

eenn

ess

cent

ralit

y

b*** ***

Degree

0 ≥0.1

Betweennesscentrality

1 ≥200

Supplementary Figure 2 Topological analysis of the meta-adhesome interaction network. (a) Clustered protein-protein interaction network model of the meta-adhesome. The largest connected graph component is displayed, comprising 11,430 interactions (black lines; edges) between 2,035 proteins (circles; nodes). Node size is proportional to degree and node colour is proportional to betweenness centrality. Black node borders indicate literature-curated adhesome4 components, which are labelled with gene names. (b) Betweenness centrality (a measure of the control a node

exerts over the interactions of other nodes in the network) for each protein is plotted according to the number of datasets in which it was identified. Box-and-whisker plot shows the median (line), mean (plus sign), 25th and 75th percentiles (box) and 5th and 95th percentiles (whiskers) (n = 1,117, 518, 238, 102, 33, 25 and 10 mapped proteins identified in 1–7 datasets, respectively, with degree ≥ 1). *P < 0.05, **P < 0.01, ***P < 0.001; Kruskal–Wallis test with Dunn’s post hoc correction (see Supplementary Table 15 for statistics source data).





1 7

Datasetoccurrence

a

b

Proteins

Biol

ogic

al p

roce

sses

Blood vessel morphogenesis Angiogenesis Vasculature development Blood vessel development Neuron projection morphogenesis Cell morphogenesis (neuron differentiation) Axonogenesis Axon guidance Integrin-mediated signalling Leukocyte migration Receptor metabolic process Interspecies interaction between organisms

ITG

A5

ITG

B3

ITG

AV

ITG

B1

CO

L1A2

C

OL3

A1

RH

OG

V

AV2

AC

TN1

VC

L T

LN1

RAC

1 S

RC

AN

XA7

GN

A13

FN

1 K

RT1

P

LAT

SER

PIN

E2

CAL

M1

ITPR

3 G

NAS

G

NB1

R

AP1A

R

AP1B

A

CTN

4 C

AP1

GN

AI2

GN

AI3

STX

4 S

TXBP

3 C

1QBP

M

MP1

C

SK

AN

XA8

CD

59

EFE

MP2

T

FPI2

T

FPI

PAB

PC4

FER

MT3

G

NA1

1 C

ALU

A

BCC

4 W

DR

1 P

RKC

B B

SG

ITG

A3

ITG

A4

YES

1 S

LC3A

2 L

YN

FG

B F

GG

T

UBA

4A

PPI

A H

GF

AN

XA5

YW

HAZ

S

OD

1 IT

GA5

IT

GB3

IT

GAV

IT

GB1

C

OL1

A2

CO

L3A1

R

HO

G

VAV

2 A

CTN

1 V

CL

TLN

1 R

AC1

SR

C

Regulation of body fluid levels Haemostasis Coagulation Blood coagulation Wound healing Platelet activation Platelet degranulation Exocytosis Cell-ECM adhesion Cell-substrate adhesion Cell junction organisation Cell junction assembly Cell-substrate junction assembly Actin filament bundle assembly Actin cytoskeleton organisation Actin filament-based process Actin filament organisation Cellular component organisation

1

2

1

2

Supplementary Figure 3 Functional enrichment map of the meta-adhesome. (a) Overrepresented biological process terms from proteins identified in the meta-adhesome were hierarchically clustered according to proteomic dataset occurrence. This identified clusters of similarly detected proteins associated

with a similar set of functional terms. (b) The two clusters containing proteins detected in the most datasets (grey boxes in a; 1, 2) are shown in detail. Proteins are labelled with gene names for clarity (see Supplementary Table 3 for details).




ACTN4 ILK ITGA5 ITGAV ITGB1 LASP1 PDLIM5 TGM2 VASP VCL FHL3 GIT2 LIMS1 LPP PALLD FHL2 ACTN1 ARHGEF7 FERMT2 TRIP6 TES PXN PTK2 PLS3 PDLIM7 PARVA LIMD1 P4HB PPIB CALD1 ITGB3 ZYX CNN2 CSK PDLIM1 SORBS1 RSU1 TLN1 FLNC TNS3 IQGAP1 FAU DDX18 BRIX1 DDX27 FBLIM1 H1FX HP1BP3 POLDIP3 SORBS3 TGFB1I1 ALYREF ANXA1 DIMT1 RPL23A SYNCRIP DNAJB1 FEN1 MRTO4 SIPA1

Occ

urre

nce

Datasets

Prot

eins

≥2<2


7

Datasetoccurrence

5

K562

MEF

A3

75

HFF

M

KF1

MKF

2

MKF

3

26 5241 51 57 59 55


LIM

dom

ain

Adhe

som

e

Consensus proteins:

Supplementary Figure 4 Comparison of IAC proteomes in the consensus adhesome. Proteins identified in the consensus adhesome were analysed by unsupervised hierarchical clustering. The binary heat map shows proteins at least two-fold enriched to FN over the negative control (red). Dataset occurrence is plotted for each protein (rainbow), literature-curated adhesome4 components are indicated by purple bars, and the presence of a LIM domain is indicated by grey bars. Dataset dissimilarity is measured by Pearson correlation. The numbers of consensus adhesome proteins

identified in each IAC proteome are displayed below the heat map. Details of the proteomic datasets are provided in Supplementary Table 1, and details of proteins identified in the consensus adhesome are provided in Supplementary Table 4. K562, human chronic myelogenous leukaemia cells11; MEF, mouse embryonic fibroblast cells (this study); A375, human malignant melanoma cells14; HFF, human foreskin fibroblast cells13; MKF1, mouse kidney fibroblast cells15; MKF2 and MKF3, mouse kidney fibroblast cells16.




ACTN4 ILK ITGA5 ITGAV ITGB1 LASP1 PDLIM5 TGM2 VASP VCL FHL3 GIT2 LIMS1 LPP PALLD FHL2 ACTN1 ARHGEF7 FERMT2 TRIP6 TES PXN PTK2 PLS3 PDLIM7 PARVA LIMD1 P4HB PPIB CALD1 ITGB3 ZYX CNN2 CSK PDLIM1 SORBS1 RSU1 TLN1 FLNC TNS3 IQGAP1 FAU DDX18 BRIX1 DDX27 FBLIM1 H1FX HP1BP3 POLDIP3 SORBS3 TGFB1I1 ALYREF ANXA1 DIMT1 RPL23A SYNCRIP DNAJB1 FEN1 MRTO4 SIPA1

Occ

urre

nce

Datasets

Prot

eins

≥2<2


7

Datasetoccurrence

5

K562

MEF

A3

75

HFF

M

KF1

MKF

2

MKF

3

26 5241 51 57 59 55


LIM

dom

ain

Adhe

som

e

Consensus proteins:


3 9 32

MYH9*PLEC*ANXA1BRIX1DIMT1PPIBMRTO4DDX18FAURPL23A

DDX27H1FXPOLDIP3HP1BP3ALYREFARHGEF2PPP2CA*SYNCRIPFEN1

MSNACTN4VIM*TUBA1B*ITGA5FLNA*MACF1EZR*CFL1

CTTNZYX

IQGAP1NUDT16L1*RAVER1*GRB2*PDLIM7TGM2CRKL*

VCLILKPDLIM5LASP1VASP

KTN1DNAJB1

ACTN1CORO2A

Time after FN-mediated cell spreading (min)

TLN1FLNCLIMS1ITGB1INPPL1*RSU1PPFIA1SIPA1CSRP1*

FHL3MAPK1*CNN2PARVB*FERMT3HSPB1*CORO1B

PDLIM1LYN*ITGAVRAC1*ITGA4*

RNA processingRibonucleoprotein complexRNA bindingSpliceosomeNucleotide binding

Actin cytoskeleton organisationNon-membrane-bounded organelleActin bindingRegulation of actin cytoskeletonActinin-type, actin-binding

Vesicle-mediated transportExtrinsic to membraneEndocytosis

Regulation of protein ubiquitinationNon-membrane-bounded organelleRNA bindingProteasomeRNA recognition motif

Membrane organisationMelanosomeActin binding

GlycolysisCitrate cycle (TCA cycle)

0.82

0.86

0.87

0.91

0.88

0.89

0.83

0.95

0 100Proportion of

maximum (%):

P4HB

RHOA*

PTPN1*

DNM2*

ARPC2*

Consensus and curated adhesome

proteins Selected functional annotations

SA1

SA2

SA3

SA4

SA5

SA6

SA7

SA8

SA9

SA10

SA11

SA12

Cluster

0.93

0.89

0.89

0.92

CytosolProteasome component region PCI

Cytoskeletal protein binding

0

100

0 10 20 30

50

Time (min)

Prop

ortio

n of

max

imum

(%)

0

100

0 10 20 30

50

0

100

0 10 20 30

50

0

100

0 10 20 30

50

0

100

0 10 20 30

50

0

100

0 10 20 30

50

0

100

0 10 20 30

50

0

100

0 10 20 30

50

Clustermean

Supplementary Figure 5 Hierarchical clustering analysis of meta-adhesome proteins identified during IAC assembly. IACs were isolated from K562 cells in biological duplicate after 3, 9 and 32 min incubation with FN-coated beads and analysed by MS (data are from 2 independent experiments; see Supplementary Table 11). Throughout IAC maturation, 1,266 of the 2,412 meta-adhesome proteins were identified and were analysed by unsupervised hierarchical clustering, revealing distinct temporal profiles of protein recruitment to IACs. Quantitative heat map displays mean spectral counts as a proportion of the maximum spectral count for each given protein. Twelve clusters were chosen on the basis of a Pearson correlation threshold greater than 0.8, labelled SA1–12, and are indicated by blue

and green bars. Literature-curated adhesome4 and consensus adhesome proteins identified in each cluster are indicated by gene name (italic, literature-curated adhesome; regular, consensus adhesome; bold, literature-curated adhesome and consensus adhesome). Literature-curated adhesome proteins that interact with consensus adhesome molecules in interaction network analyses are indicated by an asterisk (see Supplementary Table 7 for details). Clusters are shown alongside corresponding profile plots, with the mean temporal profile for each cluster indicated by a red line. The most significantly overrepresented functional annotations for selected clusters are listed. Full details of enriched functional terms are provided in Supplementary Table 13.





ENAH*

IQGAP1ACTN4ACTN1LPPCSRP1*CTTNTLN1

VIM*TNS3VCLFLNCMYH9*VASPITGA3*

PTK2FHL2ITGB5*

TUBA1B*FEN1ANXA1ITGA5FAUP4HBH1FXPPIBTGM2

SLC3A2*PLEC*FLNA*ITGB3ITGAVITGB1RPL23AFBLIM1RHOA*

CFL1

FHL3ILK

SORBS1

TUBA4A

0 5 10 15

Time after nocodazole

washout (min)

0 100Proportion of

maximum (%):

PLS3TGFB1I1TRIP6ZYXPPFIA1

PDLIM7PALLDFERMT2LASP1RAC1*SVIL*

LIMD1HSPB1*PARVACALD1TNS1*

SD10

0.92

0.90

0.89

0.84

0.81

0.83

0.810.86

Consensus and curated adhesome

proteins

ALYREF

SYNCRIP

CAV1*

0

50

100

0 5 10 15

MACF1PDLIM5CNN2PXN

EZR*MSNRDX

PRNP*CORO1B

ARPC2*

GIT1*

0.86

0.83

0.92

0.89

0.91

Selected functional annotations

Actin-filament based processActin cytoskeletonActin bindingTight junctionActinin-type, actin-binding

Cytoskeleton organisationCytoskeletonActin bindingZinc finger, LIM-type

Cytoskeleton organisationActin cytoskeletonCytoskeletal protein binding

Translation elongationCytosolic ribosomeStructural molecule activityRibosomeCollagen triple helix repeat

SD1

SD2

SD3

SD5

SD6SD7SD8SD9

SD4

SD12

SD11

SD13

SD14

SD15

SD17SD16

Cluster

0.970.91

0.880.91

CytoskeletonActin bindingRegulation of actin cytoskeletonEzrin/radixin/moesin

GTPase activity

Time (min)

Prop

ortio

n of

max

imum

(%)

0

50

100

0 5 10 15

0

50

100

0 5 10 15

0

50

100

0 5 10 15

0

50

100

0 5 10 15

0

50

100

0 5 10 15

0

50

100

0 5 10 15

0

50

100

0 5 10 15

Adhesion assembly dataset

Meta-adhesome

Meta-adhesome

195717

45543

188

Adhesion disassembly

dataset

115321

974 126639

b

c

a

Clustermean

Supplementary Figure 6 Hierarchical clustering analysis of meta-adhesome proteins identified during IAC disassembly. (a) IACs were isolated from adherent U2OS cells in biological triplicate upon nocodazole removal and 5, 10 and 15 min after nocodazole washout to examine changes in IAC composition throughout IAC disruption32. Isolated IACs at each time point were analysed by MS (data are from 3 independent experiments; see Supplementary Table 12). Throughout IAC disassembly, 455 of the 2,412 meta-adhesome proteins were identified and were analysed by unsupervised hierarchical clustering, revealing distinct temporal profiles of protein dissociation from IACs. Quantitative heat map displays mean spectral counts as a proportion of the maximum spectral count for each given protein. Seventeen clusters were chosen on the basis of a Pearson correlation threshold greater than 0.8, labelled SD1–17, and are indicated by blue and green bars. Literature-curated adhesome4 and consensus

adhesome proteins identified in each cluster are indicated by gene name (italic, literature-curated adhesome; regular, consensus adhesome; bold, literature-curated adhesome and consensus adhesome). Literature-curated adhesome proteins that interact with consensus adhesome molecules in interaction network analyses are indicated by an asterisk (see Supplementary Table 7 for details). Clusters are shown alongside corresponding profile plots, with the mean temporal profile for each cluster indicated by a red line. The most significantly overrepresented functional annotations for selected clusters are listed. Full details of enriched functional terms are provided in Supplementary Table 14. (b,c) Area-proportional Venn diagrams showing the overlap between the meta-adhesome and proteins identified by MS during IAC assembly (b) or IAC disassembly (c). For each set, the total number of proteins (black text) and the number of proteins identified in the consensus adhesome (bold red text) is indicated.





aDMSO Nocodazole 0 10 305 15

Paxillin

*****

b c

*****

***

0

5

10

15

20

25

0

5

10

15

0 3015105Noc

DMSO 0 3015105

Pos

itive

are

a (p

ropo

rtion

of

tota

l cel

l are

a, %

)

β1 integrin**

ed**

**

0

2

4

6

8

10

0

5

10

15

20

25

0 3015105 0 3015105

Time after nocodazole washout (min)

Noc

DMSONoc

DMSONoc

DMSOTime after Nocwashout (min)

Time after Nocwashout (min)



*

*******

** ******

***

pPax

illin

Y11

8pF

AK

Y39

7P

axill

inβ1

inte

grin

pPaxillinY118

pFAKY397

Supplementary Figure 7 Changes in additional consensus adhesome components during IAC disassembly. (a) To examine IAC dynamics during microtubule-induced IAC disassembly32, HFF cells treated with DMSO, 10 µM nocodazole or after nocodazole removal at different times were stained for phospho-paxillinY118, paxillin, phospho-FAKY397 and β1 integrin. Representative images are shown. Scale bars, 20 µm. (b–e) Quantification of images in a. Phospho-paxillinY118 (b), paxillin (c), phospho-FAKY397 (d)

and β1 integrin (e) levels were quantified as a proportion of total cell area. Box-and-whisker plots show median (line), mean (plus sign), 25th and 75th percentiles (box) and 5th and 95th percentiles (whiskers) (n = 10 cells per condition from one independent experiment). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; Kruskal–Wallis test with Dunn’s post hoc correction (see Supplementary Table 15 for statistics source data). Noc, nocodazole.




Supplementary Table Legends

Supplementary Table 1 Proteomic datasets assembled in the meta-adhesome. Seven datasets detailing the composition of FN-induced IACs were assembled to create the meta-adhesome database. Metadata for each dataset are provided.

Supplementary Table 2 Proteins identified in purified mouse embryonic fibroblast integrin adhesion complexes by mass spectrometry. IACs were isolated from MEF cells spread on FN or transferrin (as a control) for 2 h. All proteins identified by mass spectrometry are detailed.

Supplementary Table 3 The meta-adhesome. Proteins at least two-fold enriched to FN-induced IACs over a corresponding negative control ligand condition in at least one of seven datasets were incorporated into the meta-adhesome database. The database contains 2,412 proteins identified in at least one dataset.

Supplementary Table 4 The consensus adhesome. Proteins enriched in at least five proteomic datasets in the meta-adhesome database were incorporated into the consensus adhesome, excluding ECM or secreted proteins. The consensus adhesome contains 60 proteins commonly identified in IAC proteomic datasets, and functional information for each protein is provided.

Supplementary Table 5 Functional enrichment analysis of the consensus adhesome. Consensus adhesome proteins were subjected to functional enrichment analysis against terms from the Gene Ontology (biological process, cellular component, molecular function), KEGG pathways and InterPro protein domains. Significantly overrepresented terms are indicated.

Supplementary Table 6 Reported interactions between consensus adhesome proteins. Evidence for protein-protein interactions between proteins in the consensus adhesome was manually verified and scored. A list of all reported interactions and corresponding source publications is provided.

Supplementary Table 7 Consensus adhesome protein binding partners of non-consensus meta-adhesome proteins. Proteins identified from the meta-adhesome that interact with consensus adhesome proteins in network analyses (Fig. 1d), termed consensus interactors, are indicated.

Supplementary Table 8 Proteins identified in purified mouse embryonic fibroblast integrin adhesion complexes, in the presence or absence of myosin II inhibition, by mass spectrometry. IACs were isolated from MEF cells spread on FN or transferrin (as a control) for 2 h, in the presence or absence of 50 µM blebbistatin, in biological duplicate. All proteins identified by mass spectrometry are detailed.

Supplementary Table 9 Effects of myosin II inhibition on integrin adhesion complex composition. (a,b) To analyse effects of myosin II inhibition on IAC composition, proteins from the consensus adhesome (a) and the literature-curated adhesome (b) that were identified in at least one of three studies that analysed IAC proteomes upon blebbistatin treatment are indicated.

Supplementary Table 10 Adhesome components identified in other adhesion protein datasets. (a,b) Proteins from the consensus adhesome (a) or the literature-curated adhesome (b) that were also identified in a dataset of proteins that co-immunoprecipitated with paxillin, vinculin or talin31, a dataset of invadopodia proteins29 or a dataset of podosome proteins30 are indicated.

Supplementary Table 11 Proteins identified during integrin adhesion complex assembly by mass spectrometry. IACs were isolated from K562 cells in biological duplicate after 3, 9 and 32 min incubation with FN-coated beads. All proteins identified by mass spectrometry are detailed.

Supplementary Table 12 Proteins identified during integrin adhesion complex disassembly by mass spectrometry. IACs were isolated from U2OS cells in biological triplicate upon nocodazole removal and 5, 10 and 15 min after nocodazole washout to examine changes in IAC composition throughout IAC disruption32. All proteins identified by mass spectrometry are detailed.

Supplementary Table 13 Functional enrichment analysis of meta-adhesome proteins co-clustered during integrin adhesion complex assembly. Meta-adhesome proteins that co-clustered in hierarchical clustering analyses of IAC assembly (Supplementary Fig. 5) were subjected to functional enrichment analysis against terms from the Gene Ontology (biological process, cellular component, molecular function), KEGG pathways and InterPro protein domains. Significantly overrepresented terms are indicated.

Supplementary Table 14 Functional enrichment analysis of meta-adhesome proteins co-clustered during integrin adhesion complex disassembly. Meta-adhesome proteins that co-clustered in hierarchical clustering analyses of IAC disassembly (Supplementary Fig. 6) were subjected to functional enrichment analysis against terms from the Gene Ontology (biological process, cellular component, molecular function), KEGG pathways and InterPro protein domains. Significantly overrepresented terms are indicated.

Supplementary Table 15 Statistics source data. (a–j) Statistics source data are provided for topological analysis of the meta-adhesome interaction network (Fig. 1f, Supplementary Fig. 2b) (a,b), quantification of adhesion protein colocalisation for Rsu-1 and caldesmon (Fig. 5) (c) and quantification of cell adhesion area during nocodazole washout (Fig. 8, Supplementary Fig. 7) (d–j).


ulementary inormation - media.nature.com · 215 m l. a . ulementary inormation. 2. ... tuba4a ppia...

Documents