Expanded View Figures

Molecular Systems Biology

Quantitative protein network dynamics

Albi Celaj et al

Expanded View Figures Figure EV1. Construction of a genome-scale pool of barcoded PCA strains. A Haploid strains expressing mDHFR-fragment (F[1,2] or F[3])-tagged proteins of interest were mated and diploids were selected. Interaction between the two proteins was then verified by individually growing each diploid strain in liquid minimal media supplemented with methotrexate (MTX) for 3 days. The F[1,2]-containing haploids (from verified interactions) were then transformed with unique TagModule cassettes, mated with their matching F[3]-containing partner strains, and combined to create a pool (depicted by differently colored yeast cells). B Growth rate of all 2,394 PCA strains initially constructed and measured as the area under the OD curve (see Materials and Methods) after 75 h of growth in selective media. A cutoff of 8 was applied to identify strains that grew under MTX selection (n = 1,701). C Relating strain growth [log10(AUC)] to the cellular concentration of the least abundant protein [log10(PPM) – values taken from PaxDB] in each protein–protein interaction pair. Each dot represents one strain, and the line of best fit is shown in red. D Abundance of proteins involved in previously published PCA complexes (Tarassov et al, 2008) that were confirmed (YES) or not confirmed (NO) to grow under the conditions in this study. Protein abundance values were retrieved from PaxDB (Wang et al, 2012) and log10-transformed. In the plot, for each strain representing a protein complex, the value of the less abundant protein in that interaction was used. Box boundaries indicate 1st and 3rd quartiles, with the central line indicating median. Bottom and top whiskers extend for either 1.5 times the interquartile range or to the most distal data point, whichever yields the shortest whisker. E Paired scatterplots comparing raw fluorescence intensities of the six control samples (DMSO) in selective media. Numbers in the lower half indicate the Pearson’s r for each pairwise comparison. F Scatterplot comparing the growth rate of each strain as measured by barcode abundance following competitive growth (x-axis), to that measured in isogenic culture (y-axis). The x-axis represents log2-transformed ratio of normalized fluorescence following selective and non-selective growth (log2(+MTX/ MTX)).

EV1

Molecular Systems Biology 13: 934 | 2017

ª 2017 The Authors

▸

Albi Celaj et al



A

Mating Selection

B

Verify PPI

120

1701 PCA-PPI strains with AUC > 8

F[1,2]

Area under curve

100 F[3]

Mating Selection

F[1,2] with BC

80 60 40 20 0

F[3]

0

500

1000

1500

2000

2500

Index of PCA-PPI strains

C

D log10 of minimum protein abundance

Isogenic culture [log10 AUC]

2.0

r=0.31 1.8

1.6

1.4

1.2

1.0

−1

0

1

2

3

2

1

0

−1

3

NO

Minimum abundance [log10 ppm]

YES

validated in this study

E

F

120

Fluorescence signal

Isogenic culture [AUC]

r=0.69 100 80 60 40 20 0 −2

−1

0

1

2

3

Pooled culture [log2(+MTX/-MTX)]

Fluorescence signal Figure EV1.

ª 2017 The Authors


EV2


0.4

0.2

0.0

−0.2

−0.4

0 −0.1 −0.2 −0.3 −0.4

0.4

0.2

0.0

−0.4

−0.6

1.0

−0.5

0.0

−0.5

Ratio

NaCl

0.0

Fmp45:Hxt5 Fmp45:Hxt1 Slc1:Sna3 (*) Sna3:Rer2 (*) Pst2:Ftr1 Ftr1:Fet3

−0.2 −0.4 −0.6 −0.8

2.0

Fmp45:Hxt5 (*) Fmp45:Hxt1 (*) Slc1:Sna3 Sna3:Rer2 Pst2:Ftr1 Ftr1:Fet3

1.5 1.0 0.5 0.0 −0.5 −1.0

−1.0

0 26

−0.6 Relative growth

−0.6

0.1

Ratio

Relative growth

Relative growth

−0.2 −0.4

Ratio

atorvastatin

0

1.0

−0.5

0.5

0.0

Relative growth

Ratio

B

0

−0.8 −1 −1.5

1

0

−1

−2

−3

−1.0

−0.6 −0.8 −1

2

0.5

−0.8

0.0

−0.6

0 −0.2 −0.4

atorvastatin − MTX

0.3 0.2

4

−0.5

−0.4

Ratio

NaCl − MTX

6

4 2

−1.0

−0.2

−0.2 −0.3 −0.4

Ratio

FK506 − MTX

6

doxorubicin − MTX

0.1 0 −0.1

−0.5

Ratio

Relative growth

Relative growth

−0.5

−1.5

−1.0

1

0

−1

−2

−3

Ratio

0.0

−0.4 −0.6 −0.8 −1

−1

−1.0

Relative growth

−0.6 −0.8

0.5

−0.8

−0.4

0.5

−0.6

−0.2

atorvastatin + MTX

0.3 0.2

2 0 −0.2

0.0

−0.4

Relative growth

−0.2

NaCl + MTX

6 4

4 2 0

−0.2

FK506 + MTX

6

doxorubicin + MTX

0.0

Relative growth

Relative growth

A

Albi Celaj et al


2

105 419 1675 6700

8

32

125 500

[mM]

[nM]

H2O2

doxorubicin

Relative growth

1.5 1.0 0.5 0.0 −0.5 −1.0 0

0.6

2.4

9.6

38

152

vol %/1000

0.0

Relative growth

Fmp45:Hxt5 Fmp45:Hxt1 Slc1:Sna3 Sna3:Rer2 Pst2:Ftr1 (*) Ftr1:Fet3 (*)

2.0

Ubc13:Mms2 (*) Kti11:Elp2 Nup49:Nup57 Rad51:Rad51 (*) Cap2:Cap2

−0.2 −0.4 −0.6 −0.8 −1.0 0

10

20

30

40

50

[uM]

Figure EV2. Confirmation of pool data with isogenic culture. A Multiple strains that displayed enhanced or diminished growth (green and red, respectively) in the pool screen in doxorubicin, FK506, NaCl, or atorvastatin, as well as several control strains (black), were grown in isogenic cultures. Relative growth values were calculated for either 300 OD measurements (doxorubicin) or 150 OD measurements (FK506, atorvastatin, and NaCl) and compared to ratios derived from the pooled experiment. B Several PCA strains (indicated in the legend) with enhanced or diminished growth rates identified in the pool screen in NaCl, atorvastatin, hydrogen peroxide, and doxorubicin were grown in selective media in the presence of multiple concentrations of the four agents (indicated on the x-axis of each plot). The strains that were identified in the pool assay in each condition are marked with an asterisk. Growth relative to that in the standard environment is indicated on the y-axis.

EV3


ª 2017 The Authors

Albi Celaj et al



plasma membrane part

organic hydroxy compound transmembrane transporter activity alcohol transmembrane transporter activity doxorubicin

cation-transporting ATPase activity ion transport cation transmembrane transporter activity cation transport iron ion transport ion transmembrane transport iron ion transmembrane transport high-affinity iron ion transport

ATPase activity, coupled to transmembrane movement of ions

heat-shock

ATPase activity, coupled to transmembrane movement of substances hydrolase activity, acting on acid anhydrides, catalyzing transmembrane movement of substances menadione organic acid transport organic acid transmembrane transporter activity

substrate-specific transmembrane transporter activity substrate-specific transporter activity

H2O2

methionine

GO Term Enrichment (LOD)

transporter activity

organic acid metabolic process oxoacid metabolic process

Accumulated complexes

transmembrane transport

spermine transmembrane transporter activity spermine transport polyamine transport secondary active transmembrane transporter activity antiporter activity polyamine transmembrane transporter activity

Depleted complexes

transmembrane transporter activity

0

sorbitol

N-starvation ion transmembrane transporter activity membrane part ethanol

endoplasmic reticulum part AA-mixture

membrane

preribosome, large subunit precursor ribonucleoprotein complex biogenesis ribosome biogenesis cellular component biogenesis

single-organism transport integral component of membrane 1181-0519

plasma membrane

O-acyltransferase activity FK506

lipid metabolic process lipid biosynthetic process cellular lipid metabolic process

atorvastatin

lipid particle

2

active transmembrane transporter activity

drug transporter drug transmembrane transporter xenobiotic-transporting ATPase xenobiotic transporter

activity activity activity activity

NaCl

regulation regulation regulation regulation regulation regulation

of protein polymerization of anatomical structure size of actin polymerization or depolymerization of cellular component size of actin filament polymerization of actin filament length

Arp2/3 protein complex organic acid biosynthetic process carboxylic acid biosynthetic process

Figure EV3. Expanded GO term enrichment. Network illustrating functional trends of genes participating in dynamic complexes. For each condition, we assessed functional enrichment among genes participating in either accumulated complexes (blue) or depleted complexes (red). Lines connect terms to their respective condition(s) and are shaded by degree of GO term enrichment measured using the log-odds score (LOD).

ª 2017 The Authors


EV4



A

B

C

D

Albi Celaj et al

E

Figure EV4. Accuracy of mRNA-based protein complex predictions. A–C Protein complex abundance changes measured by BC-PCA compared to mass-action-predicted complex abundance changes based on mRNA expression data under the respiratory growth condition (i.e., in ethanol-containing medium). Each dot represents a protein complex. Predictions were performed using mRNA expression ratios at 0.5 h (A), 1 h (B), and 12 h (C) hours after the shift from glucose to ethanol-containing media. D, E Distribution of estimates of the proportion of protein complex changes which were explained by mRNA-based mass action predictions for all interactions (D) and interactions found to be significant under ethanol in BC-PCA (E). The red solid line represents the median value, and the red dashed lines represent the 5th and 95th percentile values from 4,950 combinations of “noise-added” randomizations.

EV5


ª 2017 The Authors

Albi Celaj et al

Erg25 (repressor)

A

Log2(R)



Hnm1 (repressor)

Tpo1 (repressor)

Rbd2 (repressor)

2

2

2

2

1

1

1

1

0

0

0

0

-1

-1

-1

-1

-2

-2

-2

-3

-3

-3

-4

-4

-4

-2

-3

-4

HNM1 RBD2 DID2 ERG25 XPT1 TPO1

B

Figure EV5. Knockdown of hub transcripts with CRISPRi. Validation of four CRISPRi guide RNAs using quantitative PCR. A Barplots showing changes in HNM1, RBD2, DID2, ERG25, XPT1, and TPO1 transcript levels following expression of catalytically inactive Cas9 fused to the transcriptional repressor Mxi1 (dCas9-Mxi1), and a guide RNA targeting the ERG25, HNM1, RBD2, or TPO1 gene locus (as indicated in the title of each plot). Data were normalized using ACT1 as a control. B Selective repression of the RBD2 locus using CRISPRi. Barplot showing changes in RBD2, TFC1, XPT1, UBC6, TAF10, and ALG9 transcript levels following expression of catalytically inactive Cas9 fused to the transcriptional repressor Mxi1 (dCas9-Mxi1), and a guide RNA targeting the RBD2 gene locus. Data were normalized using ACT1 as a control. The mean of three independent replicates is plotted, and error bars represent the standard deviation.

ª 2017 The Authors


EV6

Expanded View Figures

Expanded View Figures

Suggest Documents