Using an in Situ Proximity Ligation Assay to Systematically Profile

Subscriber access provided by BOSTON UNIV

Article

Using an in situ proximity ligation assay to systematically profile endogenous protein-protein interactions in a pathway network Tzu-Chi Chen, Kuan-Ting Lin, Chun-Houh Chen, Sheng-An Lee, Pei-Ying Lee, YuWen Liu, Yu-Lun Kuo, Feng-Sheng Wang, Jin-Mei Lai, and Chi-Ying F. Huang J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/pr5002737 • Publication Date (Web): 22 Sep 2014 Downloaded from http://pubs.acs.org on September 26, 2014

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Proteome Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Using an in situ proximity ligation assay to systematically profile endogenous protein-protein interactions in a pathway network Tzu-Chi Chen1,9, Kuan-Ting Lin2,9, Chun-Houh Chen3, Sheng-An Lee4, Pei-Ying Lee5, Yu-Wen Liu5, Yu-Lun Kuo6, Feng-Sheng Wang7, Jin-Mei Lai8, Chi-Ying F. Huang1,2,5* 1

Institute of Clinical Medicine, National Yang-Ming University, Taipei 112, Taiwan

2

Institute of Biomedical Informatics, Center for Systems and Synthetic Biology,

National Yang-Ming University, Taipei 112, Taiwan 3 Institute of Statistics, Academia Sinica, Taipei 115, Taiwan 4

Department of Information Management, Kainan University, Taoyuan 338, Taiwan Institute of Biopharmaceutical Sciences, National Yang-Ming University, Taipei 112, Taiwan

5

6

Department of Computer Science and Information Engineering, National Taiwan University, Taipei 112, Taiwan 7 Department of Chemical Engineering, National Chung Cheng University, Chiayi 621, Taiwan 8 Department of Life Science, Fu-Jen Catholic University, Taipei 242, Taiwan 9 These authors contributed equally to this study. *

corresponding author

Email addresses: TCC: [email protected] KTL: [email protected] CHC: [email protected] SAL: [email protected] PYL: [email protected] YWL: [email protected] YLK: [email protected] FSW: [email protected] JML: [email protected] CYH: [email protected]

1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract Signal transduction pathways in the cell require protein-protein interactions (PPIs) to respond to environmental cues. Diverse experimental techniques for detecting PPIs have been developed. However, the huge amount of PPI data accumulated from various sources poses a challenge with respect to data reliability. Herein, we collected ~700 primary antibodies and employed a highly sensitive and specific technique, an in situ proximity ligation assay, to investigate 1,204 endogenous PPIs in HeLa cells, and 557 PPIs of them tested positive. To overview the tested PPIs, we mapped them into 13 PPI public databases, which showed 72 % of them were annotated in the Human Protein Reference Database (HPRD) and 8 PPIs were new PPIs not in the PubMed database. Moreover, TP53, CTNNB1, AKT1, CDKN1A and CASP3 were the top 5 proteins prioritized by topology analyses of the 557 PPI network. Integration of the PPI-pathway interaction revealed that 90 PPIs were cross-talk PPIs linking 17 signaling pathways based on Reactome annotations. The top 2 connected cross-talk PPIs are MAPK3-DAPK1 and FAS-PRKCA interactions, which link 9 and 8 pathways, respectively. In summary, we established an open resource for biological modules and signaling pathway profiles, providing a foundation for comprehensive analysis of the human interactome.

Key words: protein-protein interaction, signaling pathway, in situ proximity ligation assay, Reactome, cross-talk protein-protein interaction

2


Page 2 of 37

Page 3 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction After successful completion of the human genome project, mapping the human interactome, which is predicted to comprise between 130,000 to 600,000 protein-protein interactions (PPIs), has become the next important challenge facing the biotech and pharmaceutical industries 1. However, accurate large-scale PPI profiling is one of the most challenging projects in the postgenetic era. One reason is the lack of significant and powerful tools for studying PPIs. Techniques, such as yeast two-hybrid (Y2H) assays

2, 3

and mass spectrometry

4, 5

, are commonly used for

high-throughput PPI studies, but PPIs generated from these tools are usually not high accuracy

6, 7

. Herein, we apply a newly developed tool, an in situ proximity ligation

assay (in situ PLA), for the visualization and quantification of endogenous PPIs in cells and tissues 8, 9. Biological functions and cellular transduction pathways are the consequence of PPIs at the cellular and molecular level. Classification of PPIs into subnetworks and signaling pathways provides a tool for understanding the molecular mechanisms and regulated modules in diseases

10-13

. To date, only a few databases use computational

analysis to map PPIs onto signaling pathways

14

. One of the limitations in studying

the interactome for cellular signaling pathways is that analysis is difficult due to the large number of PPIs with numerous specific functions or dynamic pathway changes

3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 37

(e.g., one protein may participate in several signaling pathway simultaneously). Moreover, many signaling pathways are interconnected and are integral to a complicated signaling map for basic cellular functions

12, 15, 16

. Therefore, this

complexity is an obstacle for researchers in generating a large-scale profile of biological functions and pathways for each PPI. High-confidence PPIs provide clues for examining and prioritizing proteins within specific pathways and cross-talk pathways through pathway analysis. In this study, we provide proof of concept for an integrated approach to mapping large-scale PPIs onto signaling pathways in an open resource on the internet for human interactome studies. We have identified 557 endogenous PPIs via in situ PLA using approximately 700 primary antibodies in HeLa cells and generated a PPI and pathway map. To comprehensively analyze these PPIs, a topology analysis of the PPI network and signaling pathways was applied to profile the relationship between the large number of proteins and pathways. The protein interactions from this study have deposited in the IMEx (http://www.imexconsortium.org) consortium through IntAct 17 and assigned the identifier IM-18707. The data from this study are also available on POINeT (http://poinet.bioinformatics.tw). We believe that this study will provide a fundamental strategy for investigating the PPIs, for example, the cross-talk PPIs, MAPK3-DAPK1 and FAS-PRKCA implicated in key human signaling pathways. 4


Page 5 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Methods Selection of protein-protein interaction (PPI) for validation We previously established POINeT database (http://poinet.bioinformatics.tw/), which integrated 11 publicly accessible PPI datasets and collected ~80,000 human PPIs

18

with the gene expression profile of GNF 19 and NCI60 20. From the POINeT database, we derived a list of PPIs whose genes have pathway annotations in the KEGG database (Kyoto Encyclopedia of Genes and Genomes, http://www.genome.jp/kegg/) 21

. Next, to investigate the roles of the PPIs in HeLa cells, we chose Reatome database

22

for further pathway analysis. We mapped the genes of our PPI list to antibodies

provided by Abnova Corporation and filtered out antibody-pairs from the same host species for in situ proximity ligation assay (in situ PLA). Due to budget limitations, 1,204 of 2,786 PPIs were tested in HeLa cells via in situ PLA, and Abnova Corporation made the antibody-pairs of positive PPIs commercially available after our screening.

Detection of PPIs by in situ proximity ligation assay (in situ PLA). We used in situ PLA, which developed recently to monitor PPIs and protein phosphorylation with highly specificity and sensitivity 23. Ideally, we need to find the titration for each antibody without signal dots in the negative slide, which will only 5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

add one antibody for in situ PLA assay, and then use this antibody titration to validate 1,204 PPIs in our project. If there are signal dots from in situ PLA with that antibody titration, it would indicate positive PPI. However, it is difficult to do this antibody titration test in large-scale screening. Therefore, we fixed the antibody titration for each PPI detection. In our preliminary test, the rabbit probe of in situ PLA kit had a higher binding sensitivity than that of the mouse probe, so there were different antibody titrations from mouse (1:50) and rabbit (1:1200) for in situ PLA assay in our screening. The cells were washed with PBS and fixed in 3% paraformaldehyde for 30 min on ice. After washing with PBS, the cells permeabilized with 0.2% Triton X-100 in PBS for 3 min at room temperature. To reduce the non-specific signal, the cells were incubated with a blocking solution (OLINK Bioscience) for 30 min at 37 °C. Then, primary 1X antibody Diluent (OLINK Bioscience) with two primary antibodies (one mouse monoclonal antibody at 1:50 dilution and one rabbit polyclonal antibody at 1:1200 dilution) was added to the cells and incubated overnight at 4 °C. All of the procedures were performed according to the manufacturer’s instructions. The images of the cells were acquired using an Olympus BX61 microscope (Olympus) and then analyzed with image analysis software, Blob Finder V3.2

24

, which automatically counts the

number of spots per cell. We established two-step criteria, comprising the signal dots 6


Page 6 of 37

Page 7 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


per cell in the negative slide, which only add one primary rabbit antibody (named Sneg/Cneg), and signal ratio (SR): (Spos/Cpos)/(Sneg/Cneg), to normalize the antibody titration issue (Spos/Cpos is the average of signal dots per cell in PPI detection). The PPI was determined to be a positive PPI only if SR >20 and Spos/Cpos >10. The antibody-pairs for the in situ PLA for detection of the 557 PPIs are sponsored by Abnova Corporation and listed in supplementary Table 7. The characteristics of antibodies used in this study can be viewed on the Abnova Corporation website

Competition assay for in situ proximity ligation assay For the competition assay, the PPI was detected by in situ PLA with the antibodies, which were pre-incubated with various concentrations of recombinant proteins (0, 10 and 100 µg). The peptide sequences of recombinant proteins are as follows: 1. TRAF2 (AAH43492): MAAASVTPPGSLELLQPGFSKTLLGTKLEAKYLCSACRNVLRRPFQAQCGHR YCSFCLASILSSGPQNCAACVHEGIYEEGISILESSSAFPDNAARREVESLPAVC PSDGCTWKGTLKEYESCHEGRCPLMLTECPACKGLVRLGEKERHLEHECPER SLSCRHCRAPCCGADVKAHHEVCPKFPLTCDGCGKKKIPREKFQDHVKTCGK CRVPCRFHAIGCLETVEGEKQQEHEVQWLREHLAMLLSSVLEAKPLLGDQSH 7



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

AGSELLQRCESLEKKTATFENIVCVLNREVERVAMTAEACSRQHRLDQDKIE ALSSKVQQLERSIGLKDLAMADLEQKVLEMEASTYDGVFIWKISDFARKRQE AVAGRIPAIFSPAFYTSRYGYKMCLRIYLNGDGTGRGTHLSLFFVVMKGPND ALLRWPFNQKVTLMLLDQNNREHVIDAFRPDVTSSSFQRPVNDMNIASGCPL FCPVSKMEAKNSYVRDDAIFIKAIVDLTGL; 2. BCL2L1 (AAH19307): MSQSNRELVVDFLSYKLSQKGYSWSQFSDVEENRTEAPEGTESEMETPSAIN GNPSWHLADSPAVNGATGHSSSLDAREVIPMAAVKQALREAGDEFELRYRR AFSDLTSQLHITPGTAYQSFEQVVNELFRDGVNWGRIVAFFSFGGALCVESVD KEMQVLVSRIAAWMATYLNDHLEPWIQENGGWDTFVELYGNNAAAESRKG QERFNRWFLTGMTVAGVVLLGSLFSRK

Protein Filtering Using Centralities The analysis of node centrality characteristics in a network is an efficient method to understand the features and roles of each node. Various studies indicate that proteins (hubs) with more interactors are more critical 25-28, although the interpretations of this phenomenon differ

29

. The degree centrality (DC) is the number of edges associated

with a node, normalized to a quantity from 0 to 1 by dividing by the maximum associated edge number in the sub-network. A graph can be represented by an 8


Page 8 of 37

Page 9 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


adjacency matrix A, where Aij =1 if there is an edge between nodes i and j and 0 otherwise. The degree centrality (DC) of a node i is defined as follows: n

DCi = ∑ Aij j =1

High-degree nodes in a protein interaction network correspond to essential proteins and might be a good predictor of their biological importance

27

. The closeness

centrality (CC) can identify nodes close to other nodes in the biological network. These centrality values can also be applied to prioritize nodes in the network.

9



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 37

Results Systematic identification and overview of 557 endogenous PPIs with 13 PPI databases. For large-scale detection of endogenous PPIs, an in situ PLA

8, 9

was used in this

study. Initially, three sets of endogenous PPIs, including IKBKB-TRAF2, CYCS-BCL2L1 and BAD-BCL2L1, were used to demonstrate the specificity of in situ PLA through an antibody competition assay. Recombinant TRAF2 was pre-incubated with a TRAF2 antibody before in situ PLA was performed to evaluate the IKBKB and TRAF2 interaction (Figure 1A). The IKBKB-TRAF2 was detected as the weakest one among these 3 PPIs, with the average signal dots per cell (Spos/Cpos) of 12.25±2.35 and SR of 24.4±4.69 (Supplementary Figure 1). Using these criteria, the in situ PLA result of these 3 PPIs with competition assay was considered to be a negative signal. Therefore, we used Spos/Cpos >10 and SR >20 as the criteria to judge the PPI detection. In addition to endogenous PPI detection, the PPI between AKT1 and RAF1 via an overexpression assay was also validated by the in situ PLA (Figure 1B).

After we confirmed the in situ PLA for PPI detection in cells, we performed additional in situ PLA assay to test the criteria in 7 known negative PPIs and 4 known 10


Page 11 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


positive PPIs (Supplementary Figure 2). It showed that the negative and positive PPIs could be distinguished by the cut-off criteria. However, we could not rule out the possibility that PPIs with borderline in situ PLA signals might have had different results after optimizing the antibody titration. Next, we used the in situ PLA for a large-scale screen of endogenous PPIs in HeLa cells.

We collected ~700 primary antibodies for the use of in situ PLA. With a limited number of available paired antibodies, which must originate from two different hosts for the in situ PLA (e.g., one from mouse and the other from rabbit), 1,204 endogenous PPIs were tested in HeLa cells, and 557 of them were found to be positive PPIs. By mapping the 557 endogenous PPIs into 13 public domain PPI databases, we found that the Human Protein Reference Database (HPRD) covered most annotations (72%) for 557 endogenous PPIs (Figure 1C; more details are shown in Supplementary Table 1). Moreover, a survey of the literature revealed that 8 of 557 PPIs were novel PPIs without PubMed records (Supplementary Table 1).

Topological analysis of a PPI network To investigate the relationship between numerous PPIs, we constructed a PPI network with the 557 endogenous PPIs. A schematic overview of the PPI network is illustrated 11



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 37

in Figure 2. In the 557 PPI network, 524 PPIs with 309 entities formed a large, interconnected network (Figure 2, the big network) with a diameter of 16 and an average distance of 11.7. In a topology analysis with average scores for degree centrality (DC) and closeness centrality (CC), TP53, CTNNB1, AKT1, CDKN1A and CASP3 were the top 5 molecules (indicated in bigger nodes in Figure 2, Supplementary Table 2).

A systematic profile of 90 cross-talk PPIs in a signaling pathway network To elucidate the relationship between pathways and PPIs, we first mapped the interactors from the 557 endogenous PPIs to 11 pathway databases and identified possible cross-talk PPIs, which connect two independent sets of pathways (Figure 3). Since the map for signaling pathways is not yet complete, it is not surprising that huge variation in the number of cross-talk PPIs was found among different pathway databases. In the present study, we investigated Reactome

22

, which is currently the

updated and curated pathway database for human cancer, in order to further analyze cross-talk PPIs. From the 17 top sets of pathways in the Reactome hierarchy of Homo sapiens (Supplementary Table 3), 90 PPIs were identified as cross-talk PPIs, whose interactors are annotated in different sets of pathways (Supplementary Tables 4 and 5). For example, MAPK3-DAPK1 is the connection between the Signal transduction and 12


Page 13 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Apoptosis pathway in the Reactome database. Except for the INHBB-ACVR1C and FBXW11-WEE1 interactions, the other 88 cross-talk PPIs were highly connected in the 557 PPI network (Figure 2, red nodes), meaning that the cross-talk PPIs formed the network with other PPIs. This finding is consistent with notion that cross-talk PPIs can link different pathways together through various PPIs.

In cells, signaling pathways are believed to form a comprehensive map in different temporal and/or spatial situations. A large-scale analysis of endogenous PPIs may at least fill in missing links in signaling pathways at the cellular level. We then attempted to ascertain which pathways were connected in the pathway-pathway network through the 90 cross-talk PPIs (Figure 4A; for details see Supplementary Table 6). Most pathways formed a highly connected network. In Figure 4A, it can be seen that the “Circardian Clock” and “DNA Repair” pathways were the most independent, as they only connected to “Signaling Transduction” and “Apoptosis”. Moreover,

the

“Signaling

Transduction-Cell

Cycle”

and

“Immune

System-Hemostasis” pathway interactions were the most connected pathways, which involved 16 cross-talk PPIs (Figure 4A, pink edges).

The median number of pathways connected by cross-talk PPIs was 4.5. We chose the 13



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 37

top 30 PPIs, which can link at least 5 PPIs together, to illustrate the cross-talk PPI and pathway network (Figure 4B; for details see Supplementary Table 5). The most connected PPI is the MAPK3-DAPK1 and FAS-PRKCA interaction (Figure 4B, red edges), which links 9 and 8 pathways, respectively (blue edges). It is interesting to note that both DAPK1 and FAS interacted with the Apoptosis pathway. Furthermore, both MAPK3 and PRKCA had the highest hub degree in this pathway-PPI network, suggesting that the Apoptosis pathway might interact with 11 pathways (Signaling Transduction, Immune System, Gene Expression, etc.) via these two cross-talk PPIs in these 557 PPI-pathway network. However, we can not rule out the possibility that there was any bias in protein selection towards PPIs of interest. In summary, the results of these systematic analyses of endogenous PPI pairs within and between pathways may have potential application in drug discovery as they can be used to identify disrupting signaling pathways by targeting PPIs.

Discussion In the last decade, high-throughput measurement of molecules or PPIs became a good choice for studying biological processes and pathway analyses. Currently, many bioinformatics studies with calculations through open sources have been developed to prioritize significant pathways in specific cellular conditions or disease models 14


30

.

Page 15 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


However, the current obstacle consists of the low accuracy of high-throughput data. To our knowledge, this study is the first report that has integrated a large-scale validated PPI data set via a highly sensitive and specific tool, in situ PLA, into pathway analyses. We profiled 557 validated PPIs with cross-talk events to generate a comprehensive depiction of the PPI-pathway network. We believe this report will be a valuable source for computational researchers to study the interactome in pathways and disease models.

We have validated 1,204 PPIs in this study and 557 PPIs were tested positive. Notably, there were 647 negative PPIs (Supplementary Table 8). However, we did not test the different titers of antibody-pairs of 647 negative PPIs for in situ PLA. Therefore, we could not exclude the experimental issue of PPI detection. To clarify the negative PPIs, we have mapped the 647 negative PPIs into Negatome 31, which is a database of non-interacting proteins derived by literature mining, manual annotation and protein structure analysis. Since the 1,204 tested PPIs were derived from several publicly accessible PPI datasets, it is not surprising that only 3 PPIs: ETS1-CREBBP, DVL1-GSK3B, MAPK8-HRAS and were highlighted in the Negatome. On the other hand, in this large-scale screen, we have not entirely excluded the possibility of the

15



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

false positive detection among 557 positive PPIs. Therefore, the 8 new PPIs in this study might need further confirmation.

Identification of the ‘right’ targets is now a critical part of the process because of the cost of drug discovery. There is a strong evidence that drugs with the ability to disrupt specific PPI, such as the MDM2/p53 interaction, can inhibit the function of pathways essential to certain tumor cells

32, 33

. However, designing small molecule inhibitors to

disrupt PPIs has been challenging due to the unique features of the binding site and the difficulty in identifying drug-like chemical leads

34-36

. Model peptides have been

discovered, but there are still several challenges to develop peptide drugs 37. Moreover, knowledge of cellular signaling pathways can also be helpful for exploiting rational targets. For example, many of the early approaches to inhibit Ras function failed, but a comprehensive understanding of the pathways afforded new targets for Raf and MEK 38.

The signaling pathways in cells are believed to form a complex map. Currently, discerning cross-talk events is an urgent task for the identification and development of relevant pharmaceutical target candidates. The cross-talk PPIs, which mediated the 16


Page 16 of 37

Page 17 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


most number of pathways to form a cellular function complex, might have potential in future study (Figure 4B), such as MAPK3-DAPK1 interaction for the knowledge of apoptosis 39, FAS-PRKCA interactions for apoptosis in T-cell40.

To date, various pathway analysis websites based on biological data are useful tools for translating the list of genes into pathways, such as KEGG 21, NCI-Nature PID 41, 42

, BioCarta (http://www.biocarta.com), Wikipathways

Reactome NetPath

45

, PharmGKB

46

, Singalink

47

, SMPDB

48

43

, and HumanCyc

, INOH

44

,

49

. Although

bioinformatics has progressed towards an optimal methodology for mimicking cellular condition, it still has many challenges as a pathway analysis tool (i.e., incomplete and incorrect knowledge annotations, missing condition- and cell-specific conditions, and an inability to model dynamic responses)

14

. Different datasets

showed a distinct result even if we provided the same PPI input (Figure 3). For example, in KEGG, there were 72 cross-talk PPIs and the most connected was ARRB2-CTNNA1 linking 10 pathways, whereas in Reactome, ARRB2-CTNNA1 only linked 4 pathways. The integration of a dataset with PPIs and pathways, as demonstrated in the present study, might yield new insights into the future development of bioinformatics tools for pathway analysis. Here, the 557 endogenous PPIs with pathway profile in POINeT, which also provides the gene expression profile 17



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

in different tissue from GNF 19 and NCI60 20, will be an open source to researchers to download and analyze.

Conclusions To our knowledge, this study is the first report that has integrated a highly sensitive and specific tool, in situ PLA, for 557 endogenous PPIs into pathway analyses. We established a bioinformatics resource of 557 endogenous PPIs with cross-talk events to generate a comprehensive depiction of the PPI-pathway network and prioritize the critical targets for disease modules. We believe that this study will provide a fundamental strategy for investigating the PPIs implicated in key human signaling pathways.

Supporting Information Supplementary Figure 1. Verification of cut-off criteria of in situ PLA for PPI detection. Supplementary Figure 2. The cut-off criteria for PPI detection was confirmed by 7 known negative PPIs and 4 known positive PPIs in HeLa cells. Supplementary Table 1. The annotations of 557 PPIs in 13 public domain PPI databases. 18


Page 18 of 37

Page 19 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Supplementary Table 2. The degree and closeness centrality of 309 nodes in 557 PPI network was shown. Supplementary Table 3. 17 sets of pathways for pathway analysis. Supplementary Table 4. The pathway information of non-cross talk PPIs was shown. Supplementary Table 5. 90 cross-talk PPIs, which link at least two pathways, were listed. Supplementary Table 6. The pathway-pathway interaction(s) through 90 cross-talk PPIs were listed. Supplementary Table 7. The list of antibodies of 557 positive PPIs for in situ PLA. Supplementary Table 8. The list of 647 negative PPIs. This material is available free of charge via the Internet at http://pubs.acs.org

Corresponding Author Chi-Ying Huang Professor and chairman Institute of Biopharmaceutical Sciences National Yang-Ming University No. 155, Sec. 2, Linong St. Taipei 112 19



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Taiwan Phone: 886-228267904 Fax: 886-228224045 E-mail: [email protected]

Competing interests The authors declare that they have no competing interests.

Authors' contributions TCC did the experiment for verification of PPIs and wrote the manuscript. KTL collected and analyzed the protein-protein interaction network and SAL maintained the data in POINeT web server. CHC, YLK and FSW analyzed the topology of PPIs in pathways. PYL and YWL did the experiment for verification of PPIs and manually curated the PPIs via literature search. JML and CYH provided the concept, supervised the project, and modified the manuscript.

Acknowledgements This large-scale PPI screening project was done in collaboration with Abnova Corporation and supported by the Industry-Academic Cooperative Research Project 20


Page 20 of 37

Page 21 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


overseen by the National Science Council (NSC 99-3112-B-010-005-CC1) to CY Huang in Taiwan. We thank Abnova Corporation for providing antibody-pairs for the PPI testing. We thank Hsuan-Cheng Huang for suggestion of PPI analysis. This research was also supported by the grants from the National Science Council (NSC101-2627-B194-001) to FS Wang, (NSC 101-2627-B-030-001 and NSC 102-2627-B-030-001)

to

JM

Lai,

(NSC102-2325-B-010-011

and

NSC

102-2627-B-010 -001) to CY Huang and Center of Excellence for Cancer Research at Taipei

Veterans

General

Hospital

(DOH101-TD-C-111-007,

and

DOH102-TD-C-111-007) to CY Huang. We thank Abnova Corporation to provide antibodies and OLINK Bioscience to give the technical supports in this study.

21



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

References 1.

Stumpf, M. P.; Thorne, T.; de Silva, E.; Stewart, R.; An, H. J.; Lappe, M.; Wiuf, C.,

Estimating the size of the human interactome. Proc Natl Acad Sci U S A 2008, 105, (19), 6959-64. 2. Rual, J. F.; Venkatesan, K.; Hao, T.; Hirozane-Kishikawa, T.; Dricot, A.; Li, N.; Berriz, G. F.; Gibbons, F. D.; Dreze, M.; Ayivi-Guedehoussou, N.; Klitgord, N.; Simon, C.; Boxem, M.; Milstein, S.; Rosenberg, J.; Goldberg, D. S.; Zhang, L. V.; Wong, S. L.; Franklin, G.; Li, S.; Albala, J. S.; Lim, J.; Fraughton, C.; Llamosas, E.; Cevik, S.; Bex, C.; Lamesch, P.; Sikorski, R. S.; Vandenhaute, J.; Zoghbi, H. Y.; Smolyar, A.; Bosak, S.; Sequerra, R.; Doucette-Stamm, L.; Cusick, M. E.; Hill, D. E.; Roth, F. P.; Vidal, M., Towards a proteome-scale map of the human protein-protein interaction network. Nature 2005, 437, (7062), 1173-8. 3.

Stelzl, U.; Worm, U.; Lalowski, M.; Haenig, C.; Brembeck, F. H.; Goehler, H.;

Stroedicke, M.; Zenkner, M.; Schoenherr, A.; Koeppen, S.; Timm, J.; Mintzlaff, S.; Abraham, C.; Bock, N.; Kietzmann, S.; Goedde, A.; Toksoz, E.; Droege, A.; Krobitsch, S.; Korn, B.; Birchmeier, W.; Lehrach, H.; Wanker, E. E., A human protein-protein interaction network: a resource for annotating the proteome. Cell 2005, 122, (6), 957-68. 4.

Ewing, R. M.; Chu, P.; Elisma, F.; Li, H.; Taylor, P.; Climie, S.; McBroom-Cerajewski, 22


Page 22 of 37

Page 23 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


L.; Robinson, M. D.; O'Connor, L.; Li, M.; Taylor, R.; Dharsee, M.; Ho, Y.; Heilbut, A.; Moore, L.; Zhang, S.; Ornatsky, O.; Bukhman, Y. V.; Ethier, M.; Sheng, Y.; Vasilescu, J.; Abu-Farha, M.; Lambert, J. P.; Duewel, H. S.; Stewart, II; Kuehl, B.; Hogue, K.; Colwill, K.; Gladwish, K.; Muskat, B.; Kinach, R.; Adams, S. L.; Moran, M. F.; Morin, G. B.; Topaloglou, T.; Figeys, D., Large-scale mapping of human protein-protein interactions by mass spectrometry. Mol Syst Biol 2007, 3, 89. 5.

Hubner, N. C.; Bird, A. W.; Cox, J.; Splettstoesser, B.; Bandilla, P.; Poser, I.; Hyman,

A.; Mann, M., Quantitative proteomics combined with BAC TransgeneOmics reveals in vivo protein interactions. J Cell Biol 2010, 189, (4), 739-54. 6.

Stelzl, U.; Wanker, E. E., The value of high quality protein-protein interaction

networks for systems biology. Curr Opin Chem Biol 2006, 10, (6), 551-8. 7.

Ramirez, F.; Schlicker, A.; Assenov, Y.; Lengauer, T.; Albrecht, M., Computational

analysis of human protein interaction networks. Proteomics 2007, 7, (15), 2541-52. 8.

Söderberg, O.; Gullberg, M.; Jarvius, M.; Ridderstråle, K.; Leuchowius, K. J.;

Jarvius, J.; Wester, K.; Hydbring, P.; Bahram, F.; Larsson, L. G., Direct observation of individual endogenous protein complexes in situ by proximity ligation. Nature methods 2006, 3, (12), 995-1000. 9.

Jarvius, M.; Paulsson, J.; Weibrecht, I.; Leuchowius, K.-J.; Andersson, A.-C.;

Wählby, C.; Gullberg, M.; Botling, J.; Sjöblom, T.; Markova, B.; Östman, A.; Landegren, 23



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

U.; Söderberg, O., In Situ Detection of Phosphorylated Platelet-derived Growth Factor Receptor β Using a Generalized Proximity Ligation Method. Molecular & Cellular Proteomics 2007, 6, (9), 1500-1509. 10. Barabasi, A. L.; Oltvai, Z. N., Network biology: understanding the cell's functional organization. Nat Rev Genet 2004, 5, (2), 101-13. 11. Colland, F.; Jacq, X.; Trouplin, V.; Mougin, C.; Groizeleau, C.; Hamburger, A.; Meil, A.; Wojcik, J.; Legrain, P.; Gauthier, J. M., Functional proteomics mapping of a human signaling pathway. Genome Res 2004, 14, (7), 1324-32. 12. Liu, C. H.; Chen, T. C.; Chau, G. Y.; Jan, Y. H.; Chen, C. H.; Hsu, C. N.; Lin, K. T.; Juang, Y. L.; Lu, P. J.; Cheng, H. C.; Chen, M. H.; Chang, C. F.; Ting, Y. S.; Kao, C. Y.; Hsiao, M.; Huang, C. Y., An analysis of protein-protein interactions in cross-talk pathways reveals CRKL as a novel prognostic marker in hepatocellular carcinoma. Mol Cell Proteomics 2013. 13. Tien, A. C.; Lin, M. H.; Su, L. J.; Hong, Y. R.; Cheng, T. S.; Lee, Y. C.; Lin, W. J.; Still, I. H.; Huang, C. Y., Identification of the substrates and interaction proteins of aurora kinases from a protein-protein interaction model. Mol Cell Proteomics 2004, 3, (1), 93-104. 14. Khatri, P.; Sirota, M.; Butte, A. J., Ten years of pathway analysis: current approaches and outstanding challenges. PLoS Comput Biol 2012, 8, (2), e1002375. 24


Page 24 of 37

Page 25 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


15. Kestler, H. A.; Kuhl, M., From individual Wnt pathways towards a Wnt signalling network. Philos Trans R Soc Lond B Biol Sci 2008, 363, (1495), 1333-47. 16. Sorkin, A.; von Zastrow, M., Endocytosis and signalling: intertwining molecular networks. Nat Rev Mol Cell Biol 2009, 10, (9), 609-22. 17. Aranda, B.; Achuthan, P.; Alam-Faruque, Y.; Armean, I.; Bridge, A.; Derow, C.; Feuermann, M.; Ghanbarian, A. T.; Kerrien, S.; Khadake, J.; Kerssemakers, J.; Leroy, C.; Menden, M.; Michaut, M.; Montecchi-Palazzi, L.; Neuhauser, S. N.; Orchard, S.; Perreau, V.; Roechert, B.; van Eijk, K.; Hermjakob, H., The IntAct molecular interaction database in 2010. Nucleic Acids Res 2010, 38, (Database issue), D525-31. 18. Lee, S. A.; Chan, C. H.; Chen, T. C.; Yang, C. Y.; Huang, K. C.; Tsai, C. H.; Lai, J. M.; Wang, F. S.; Kao, C. Y.; Huang, C. Y., POINeT: protein interactome with sub-network analysis and hub prioritization. BMC Bioinformatics 2009, 10, 114. 19. Su, A. I.; Wiltshire, T.; Batalov, S.; Lapp, H.; Ching, K. A.; Block, D.; Zhang, J.; Soden, R.; Hayakawa, M.; Kreiman, G.; Cooke, M. P.; Walker, J. R.; Hogenesch, J. B., A gene atlas of the mouse and human protein-encoding transcriptomes. Proc Natl Acad Sci U S A 2004, 101, (16), 6062-7. 20. Shankavaram, U. T.; Reinhold, W. C.; Nishizuka, S.; Major, S.; Morita, D.; Chary, K. K.; Reimers, M. A.; Scherf, U.; Kahn, A.; Dolginow, D.; Cossman, J.; Kaldjian, E. P.; Scudiero, D. A.; Petricoin, E.; Liotta, L.; Lee, J. K.; Weinstein, J. N., Transcript and 25



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

protein expression profiles of the NCI-60 cancer cell panel: an integromic microarray study. Mol Cancer Ther 2007, 6, (3), 820-32. 21. Kanehisa, M.; Goto, S.; Sato, Y.; Kawashima, M.; Furumichi, M.; Tanabe, M., Data, information, knowledge and principle: back to metabolism in KEGG. Nucleic Acids Res 2014, 42, (1), D199-205. 22. Croft, D., Building models using Reactome pathways as templates. Methods Mol Biol 2013, 1021, 273-83. 23. Weibrecht, I.; Leuchowius, K. J.; Clausson, C. M.; Conze, T.; Jarvius, M.; Howell, W. M.; Kamali-Moghaddam, M.; Soderberg, O., Proximity ligation assays: a recent addition to the proteomics toolbox. Expert Rev Proteomics 2010, 7, (3), 401-9. 24. Allalou, A.; Wahlby, C., BlobFinder, a tool for fluorescence microscopy image cytometry. Comput Methods Programs Biomed 2009, 94, (1), 58-65. 25. Batada, N. N.; Hurst, L. D.; Tyers, M., Evolutionary and physiological importance of hub proteins. PLoS Comput Biol 2006, 2, (7), e88. 26. Hahn, M. W.; Kern, A. D., Comparative genomics of centrality and essentiality in three eukaryotic protein-interaction networks. Mol Biol Evol 2005, 22, (4), 803-6. 27. Jeong, H.; Mason, S. P.; Barabasi, A. L.; Oltvai, Z. N., Lethality and centrality in protein networks. Nature 2001, 411, (6833), 41-2. 28. Yu, H.; Greenbaum, D.; Xin Lu, H.; Zhu, X.; Gerstein, M., Genomic analysis of 26


Page 26 of 37

Page 27 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


essentiality within protein networks. Trends Genet 2004, 20, (6), 227-31. 29. Zotenko, E.; Mestre, J.; O'Leary, D. P.; Przytycka, T. M., Why do hubs in the yeast protein interaction network tend to be essential: reexamining the connection between the network topology and essentiality. PLoS Comput Biol 2008, 4, (8), e1000140. 30. Scholtens, D.; Vidal, M.; Gentleman, R., Local modeling of global interactome networks. Bioinformatics 2005, 21, (17), 3548-57. 31. Blohm, P.; Frishman, G.; Smialowski, P.; Goebels, F.; Wachinger, B.; Ruepp, A.; Frishman, D., Negatome 2.0: a database of non-interacting proteins derived by literature mining, manual annotation and protein structure analysis. Nucleic Acids Res 2014, 42, (Database issue), D396-400. 32. Gonzalez, A. Z.; Eksterowicz, J.; Bartberger, M. D.; Beck, H. P.; Canon, J.; Chen, A.; Chow, D.; Duquette, J.; Fox, B. M.; Fu, J.; Huang, X.; Houze, J. B.; Jin, L.; Li, Y.; Li, Z.; Ling, Y.; Lo, M. C.; Long, A. M.; McGee, L. R.; McIntosh, J.; McMinn, D. L.; Oliner, J. D.; Osgood, T.; Rew, Y.; Saiki, A. Y.; Shaffer, P.; Wortman, S.; Yakowec, P.; Yan, X.; Ye, Q.; Yu, D.; Zhao, X.; Zhou, J.; Olson, S. H.; Medina, J. C.; Sun, D., Selective and potent morpholinone inhibitors of the MDM2-p53 protein-protein interaction. J Med Chem 2014, 57, (6), 2472-88. 33. Yu, S.; Qin, D.; Shangary, S.; Chen, J.; Wang, G.; Ding, K.; McEachern, D.; Qiu, S.; 27



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nikolovska-Coleska, Z.; Miller, R.; Kang, S.; Yang, D.; Wang, S., Potent and orally active small-molecule inhibitors of the MDM2-p53 interaction. J Med Chem 2009, 52, (24), 7970-3. 34. Sun, C.; Hajduk, P. J., Nuclear magnetic resonance in target profiling and compound file enhancement. Curr Opin Drug Discov Devel 2006, 9, (4), 463-70. 35. Wendt, M. D.; Sun, C.; Kunzer, A.; Sauer, D.; Sarris, K.; Hoff, E.; Yu, L.; Nettesheim, D. G.; Chen, J.; Jin, S.; Comess, K. M.; Fan, Y.; Anderson, S. N.; Isaac, B.; Olejniczak, E. T.; Hajduk, P. J.; Rosenberg, S. H.; Elmore, S. W., Discovery of a novel small molecule binding site of human survivin. Bioorg Med Chem Lett 2007, 17, (11), 3122-9. 36. Zhang, X.; Betzi, S.; Morelli, X.; Roche, P., Focused chemical libraries - design and enrichment: an example of protein-protein interaction chemical space. Future Med Chem 2014. 37. Krall, A.; Brunn, J.; Kankanala, S.; Peters, M. H., A simple contact mapping algorithm for identifying potential peptide mimetics in protein-protein interaction partners. Proteins 2014. 38. Wong, K. K., Recent developments in anti-cancer agents targeting the Ras/Raf/ MEK/ERK pathway. Recent Pat Anticancer Drug Discov 2009, 4, (1), 28-35. 39. Chen, C. H.; Wang, W. J.; Kuo, J. C.; Tsai, H. C.; Lin, J. R.; Chang, Z. F.; Chen, R. H., Bidirectional signals transduced by DAPK-ERK interaction promote the apoptotic 28


Page 28 of 37

Page 29 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


effect of DAPK. EMBO J 2005, 24, (2), 294-304. 40. Chen, C. Y.; Faller, D. V., Selective inhibition of protein kinase C isozymes by Fas ligation. J Biol Chem 1999, 274, (22), 15320-8. 41. Schaefer, C. F.; Anthony, K.; Krupa, S.; Buchoff, J.; Day, M.; Hannay, T.; Buetow, K. H., PID: the Pathway Interaction Database. Nucleic Acids Res 2009, 37, (Database issue), D674-9. 42. Joshi-Tope, G.; Vastrik, I.; Gopinath, G. R.; Matthews, L.; Schmidt, E.; Gillespie, M.; D'Eustachio, P.; Jassal, B.; Lewis, S.; Wu, G.; Birney, E.; Stein, L., The Genome Knowledgebase: a resource for biologists and bioinformaticists. Cold Spring Harb Symp Quant Biol 2003, 68, 237-43. 43. Kelder, T.; van Iersel, M. P.; Hanspers, K.; Kutmon, M.; Conklin, B. R.; Evelo, C. T.; Pico, A. R., WikiPathways: building research communities on biological pathways. Nucleic Acids Res 2012, 40, (Database issue), D1301-7. 44. Yamamoto, S.; Sakai, N.; Nakamura, H.; Fukagawa, H.; Fukuda, K.; Takagi, T., INOH: ontology-based highly structured database of signal transduction pathways. Database (Oxford) 2011, 2011, bar052. 45. Kandasamy, K.; Mohan, S. S.; Raju, R.; Keerthikumar, S.; Kumar, G. S.; Venugopal, A. K.; Telikicherla, D.; Navarro, J. D.; Mathivanan, S.; Pecquet, C.; Gollapudi, S. K.; Tattikota, S. G.; Mohan, S.; Padhukasahasram, H.; Subbannayya, Y.; Goel, R.; Jacob, H. 29



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

K.; Zhong, J.; Sekhar, R.; Nanjappa, V.; Balakrishnan, L.; Subbaiah, R.; Ramachandra, Y. L.; Rahiman, B. A.; Prasad, T. S.; Lin, J. X.; Houtman, J. C.; Desiderio, S.; Renauld, J. C.; Constantinescu, S. N.; Ohara, O.; Hirano, T.; Kubo, M.; Singh, S.; Khatri, P.; Draghici, S.; Bader, G. D.; Sander, C.; Leonard, W. J.; Pandey, A., NetPath: a public resource of curated signal transduction pathways. Genome Biol 2010, 11, (1), R3. 46. Whirl-Carrillo, M.; McDonagh, E. M.; Hebert, J. M.; Gong, L.; Sangkuhl, K.; Thorn, C. F.; Altman, R. B.; Klein, T. E., Pharmacogenomics knowledge for personalized medicine. Clin Pharmacol Ther 2012, 92, (4), 414-7. 47. Fazekas, D.; Koltai, M.; Turei, D.; Modos, D.; Palfy, M.; Dul, Z.; Zsakai, L.; Szalay-Beko, M.; Lenti, K.; Farkas, I. J.; Vellai, T.; Csermely, P.; Korcsmaros, T., SignaLink 2 - a signaling pathway resource with multi-layered regulatory networks. BMC Syst Biol 2013, 7, 7. 48. Frolkis, A.; Knox, C.; Lim, E.; Jewison, T.; Law, V.; Hau, D. D.; Liu, P.; Gautam, B.; Ly, S.; Guo, A. C.; Xia, J.; Liang, Y.; Shrivastava, S.; Wishart, D. S., SMPDB: The Small Molecule Pathway Database. Nucleic Acids Res 2010, 38, (Database issue), D480-7. 49. Romero, P.; Wagg, J.; Green, M. L.; Kaiser, D.; Krummenacker, M.; Karp, P. D., Computational prediction of human metabolic pathways from the complete human genome. Genome Biol 2005, 6, (1), R2.

30


Page 30 of 37

Page 31 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure Captions Figure 1. Detection of endogenous 1,204 PPIs via in situ PLA and mapping PPIs into 13 PPI databases. (A) To support confidence in the in situ PLA, three well-known PPIs were further confirmed through a competition assay. Prior to in situ PLA, the target antibodies (e.g., TRAF2 antibody) were first incubated with a series of recombinant proteins at varying concentrations (0, 10 and 100 µg). The interaction signal decreased with antibody competition. (B) Using the in situ PLA to detect AKT1-RAF1 interactions, we quantified the signals (blobs/cell) for endogenous and overexpressed AKT1-RAF1 interactions in HEK293T cells. AKT1-RAF1 was not detected endogenously, whereas it was highly expressed when AKT1 and RAF1 were overexpressed. (C) Percentages of 557 PPIs annotated in 13 public domain PPI databases. Lists of PPIs for different PPI databases were derived from CPDB (ConsensusPathDB). The PPI list of IntAct were downloaded from the IntAct website. From left to right, we ranked the databases by percentage from high to low. HPRD covered the most PPIs in the 557 PPIs, whereas COURM covered the least.

Figure 2. An architectural map of the network comprising the 557 endogenous PPIs. The PPI network involved in 557 PPIs with 366 nodes. Of the 557 PPIs, 524 31



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

formed the connective network with 309 nodes. The nodes with 90 cross-talk PPIs are labeled in red node and pink edge. The larger nodes are with higher hub degree.

Figure 3. Numbers of cross-talk PPIs in different pathway database annotations. Due to the variations of pathway annotations among databases, there are different numbers of cross-talk PPIs whose gene pairs are annotated in distinct pathways. For example, based on annotations in Reactome, there were 90 cross-talk PPIs in the 557 PPIs reported in the study. Also, there were common cross-talk PPIs among different databases. For example, 53 of the 90 cross-talk PPIs from Reactome were the same as those based on the annotation in PID.

Figure 4. The pathway network from cross-talk PPIs. (A) The pathway-pathway interaction network by 90 cross-talk PPIs. The edge width indicates the number of cross-talk PPIs in the connection of two pathways. The wider edge represents more cross-talk PPIs involved in the pathway connection. The node size represents the hub degree of pathway in the pathway interaction network. (B) The cross-talk PPI and pathway interaction network are shown.

32


Page 32 of 37

Page 33 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. Detection of endogenous 1,204 PPIs via in situ PLA and mapping PPIs into 13 PPI databases. (A) To support confidence in the in situ PLA, three well-known PPIs were further confirmed through a competition assay. Prior to in situ PLA, the target antibodies (e.g., TRAF2 antibody) were first incubated with a series of recombinant proteins at varying concentrations (0, 10 and 100 µg). The interaction signal decreased with antibody competition. (B) Using the in situ PLA to detect AKT1-RAF1 interactions, we quantified the signals (blobs/cell) for endogenous and overexpressed AKT1-RAF1 interactions in HEK293T cells. AKT1-RAF1 was not detected endogenously, whereas it was highly expressed when AKT1 and RAF1 were overexpressed. (C) Percentages of 557 PPIs annotated in 13 public domain PPI databases. Lists of PPIs for different PPI databases were derived from CPDB (ConsensusPathDB). The PPI list of IntAct were downloaded from the IntAct website. From left to right, we ranked the databases by percentage from high to low. HPRD covered the most PPIs in the 557 PPIs, whereas COURM covered the least. 125x107mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. An architectural map of the network comprising the 557 endogenous PPIs. The PPI network involved in 557 PPIs with 366 nodes. Of the 557 PPIs, 524 formed the connective network with 309 nodes. The nodes with 90 cross-talk PPIs are labeled in red node and pink edge. The larger nodes are with higher hub degree. 163x193mm (300 x 300 DPI)


Page 34 of 37

Page 35 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. Numbers of cross-talk PPIs in different pathway database annotations. Due to the variations of pathway annotations among databases, there are different numbers of cross-talk PPIs whose gene pairs are annotated in distinct pathways. For example, based on annotations in Reactome, there were 90 cross-talk PPIs in the 557 PPIs reported in the study. Also, there were common cross-talk PPIs among different databases. For example, 53 of the 90 cross-talk PPIs from Reactome were the same as those based on the annotation in PID. 42x18mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. The pathway network from cross-talk PPIs. (A) The pathway-pathway interaction network by 90 cross-talk PPIs. The edge width indicates the number of cross-talk PPIs in the connection of two pathways. The wider edge represents more cross-talk PPIs involved in the pathway connection. The node size represents the hub degree of pathway in the pathway interaction network. (B) The cross-talk PPI and pathway interaction network are shown. 209x297mm (300 x 300 DPI)


Page 36 of 37

Page 37 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Graphical Abstract 169x108mm (300 x 300 DPI)


Using an in Situ Proximity Ligation Assay to Systematically Profile

Recommend Documents